Solvation Keywords in the Jaguar Input File
- Overview
- Examples
Jaguar provides four continuum solvation models, PBF, SM6/SM8, SMD, and PCM. The solvation keywords that apply to all models or to more than one model are described in Table 1. By default, Jaguar calculations are performed in the gas phase, so isolv=0 and all other solvation keywords are irrelevant. Keywords for each of the solvation models are given in the sections below.
For solvated geometry optimizations, the trust keyword, which is described in Geometry Optimization and Transition-State Keywords in the Jaguar Input File, has a default value of 0.1 instead of its usual default of 0.3.
Note: QSite does not support continuum solvation, so you cannot add these keywords to a QSite input file.
|
Keyword |
Value |
Description |
|
0 |
Do not perform a solvation calculation. |
|
|
|
2 |
Perform a solvation calculation using a Poisson-Boltzmann solver [15, 222] |
|
|
5 |
Perform a solvation calculation using the SM6 model [231], which applies only to water. This is the only solvation keyword that should be specified when performing SM6 calculations. Only available for single-point calculations. |
|
|
6 |
Perform a solvation calculation using the SM8 model [223]. Can be used with the solvent keyword for any of the recognized solvents; other keywords are not relevant. Only available for single-point calculations. |
|
|
7 |
Perform a solvation calculation using the polarizable continuum model (PCM). The specific variant of the model is set with the pcm_model keyword. |
|
|
8 |
Perform a solvation calculation using the SMD model [321]. Can be used with the solvent keyword for any of the recognized solvents; other keywords are not relevant. The SMD solvent model is similar to SM8 with a re-fit non-electrostatic contribution to the free energy and the use of CPCM for the electrostatic contribution (rather than generalized Born like in SM8). |
| solvent_model |
gas vacuum pbf sm6 sm8 pcm smd |
Sets isolv = 0 Sets isolv = 0 Sets isolv= 2 Sets isolv = 5 Sets isolv = 6 Sets isolv = 7 Sets isolv = 8 |
|
water |
Solvent name. The allowed values are listed below. |
|
|
0 |
The ideal gas entropy approximation is used for the calculation of the entropy and total free energy. |
|
|
|
1 |
Garza's Sω entropy approximation is used for the calculation of the entropy and total free energy. See Solvation Entropy in Jaguar. |
|
|
2 |
Garza's Sε entropy approximation is used for the calculation of the entropy and total free energy. See Solvation Entropy in Jaguar. |
|
|
3 |
Garza's Sεα entropy approximation is used for the calculation of the entropy and total free energy. See Solvation Entropy in Jaguar. |
|
0 |
Do not run a single-point PBF calculation after optimization with PCM solvation. |
|
|
|
1 |
Run a single-point PBF calculation after optimization with PCM solvation. |
|
0 |
Do not run a single-point PBF calculation after optimization with SMD solvation. |
|
|
|
1 |
Run a single-point PBF calculation after optimization with SMD solvation. |
|
80.37 |
Dielectric constant of solvent. Used for PBF, PCM, and SM8 models. |
|
|
1.5×10−4 |
Solvation energy convergence criterion in Hartrees |
|
|
number |
Gas phase energy of molecule, in Hartrees; used in some restart (new input) files for solvation jobs |
Available Solvents
The solvents available for use with the solvent keyword with the PBF, PCM, and SM8 models are listed below. Solvents are defined by their dielectric constant and their optical dielectric constant.
|
1,1,1-trichloroethane |
acetonitrile |
fluorobenzene |
|
1,1,2-trichloroethane |
acetophenone |
formamide |
|
1,2,4-trimethylbenzene |
aniline |
formic_acid |
|
1,2-dibromoethane |
anisole |
n-heptane |
|
1,2-dichloroethane |
benzaldehyde |
n-hexadecane |
|
1,2-ethanediol |
benzene |
n-hexane |
|
1,4-dioxane |
benzonitrile |
hexanoic_acid |
|
1-bromo-2-methylpropane |
benzyl_alcohol |
iodobenzene |
|
1-bromooctane |
bromobenzene |
iodoethane |
|
1-bromopentane |
bromoethane |
iodomethane |
|
1-bromopropane |
bromoform |
isopropylbenzene |
|
1-butanol |
butanal |
p-isopropyltoluene |
|
1-chlorohexane |
butanoic_acid |
mesitylene |
|
1-chloropentane |
butanone |
methanol |
|
1-chloropropane |
butanonitrile |
methyl_benzoate |
|
1-decanol |
butyl_ethanoate |
methyl_butanoate |
|
1-fluorooctane |
butylamine |
methyl_ethanoate |
|
1-heptanol |
n-butylbenzene |
methyl_methanoate |
|
1-hexanol |
sec-butylbenzene |
methyl_propanoate |
|
1-hexene |
tert-butylbenzene |
N-methylaniline |
|
1-hexyne |
carbon_disulfide |
methylcyclohexane |
|
1-iodobutane |
carbon_tetrachloride |
N-methylformamide_(E/Z_mixture) |
|
1-iodohexadecane |
chlorobenzene |
nitrobenzene |
|
1-iodopentane |
chloroform |
nitroethane |
|
1-iodopropane |
a-chlorotoluene |
nitromethane |
|
1-nitropropane |
o-chlorotoluene |
o-nitrotoluene |
|
1-nonanol |
m-cresol |
n-nonane |
|
1-octanol |
o-cresol |
n-octane |
|
1-pentanol |
cyclohexane |
n-pentadecane |
|
1-pentene |
cyclohexanone |
pentanal |
|
1-pentyne |
cyclopentane |
n-pentane |
|
1-propanol |
cyclopentanol |
pentanoic_acid |
|
2,2,2-trifluoroethanol |
cyclopentanone |
pentyl_ethanoate |
|
2,2,4-trimethylpentane |
decalin_(cis/trans_mixture) |
pentylamine |
|
2,4-dimethylpentane |
cis-decalin |
perfluorobenzene |
|
2,4-dimethylpyridine |
n-decane |
propanal |
|
2,6-dimethylpyridine |
dibromomethane |
propanoic_acid |
|
2-bromopropane |
dibutyl_ether |
propanonitrile |
|
2-butanol |
o-dichlorobenzene |
propyl_ethanoate |
|
2-chlorobutane |
E-1,2-dichloroethene |
propylamine |
|
2-heptanone |
Z-1,2-dichloroethene |
pyrrolidine |
|
2-hexanone |
dichloromethane |
pyridine |
|
2-methoxyethanol |
diethyl_ether |
tetrachloroethene |
|
2-methyl-1-propanol |
diethyl_sulfide |
tetrahydrofuran |
|
2-methyl-2-propanol |
diethylamine |
tetrahydrothiophene-S,S-dioxide |
|
2-methylpentane |
diiodomethane |
tetralin |
|
2-methylpyridine |
diisopropyl_ether |
thiophene |
|
2-nitropropane |
cis-1,2-dimethylcyclohexane |
thiophenol |
|
2-octanone |
dimethyl_disulfide |
toluene |
|
2-pentanone |
N,N-dimethylacetamide |
trans-decalin |
|
2-propanol |
N,N-dimethylformamide |
tributyl_phosphate |
|
2-propen-1-ol |
dimethylsulfoxide |
trichloroethene |
|
E-2-pentene |
diphenyl_ether |
triethylamine |
|
3-methylpyridine |
dipropylamine |
n-undecane |
|
3-pentanone |
n-dodecane |
water |
|
4-heptanone |
ethanethiol |
xylene_(mixture) |
|
4-methyl-2-pentanone |
ethanol |
m-xylene |
|
4-methylpyridine |
ethyl_ethanoate |
o-xylene |
|
5-nonanone |
ethyl_methanoate |
p-xylene |
|
acetic_acid |
ethyl_phenyl_ether |
|
|
acetone |
ethylbenzene |
|
PBF Model
Most of the PBF solvation keywords correspond to GUI options described in Jaguar Solvation Settings. The keywords for the PBF model are described in Table 2. Defaults for some of these keywords are not indicated in bold italics, since the keywords’ default values generally depend on other keywords. The default values for the real-valued parameters are for water. The solvent can be specified with the solvent keyword. For a custom solvent, you can specify the dielectric constant and probe radius; the solvent keyword sets these two quantities as described below.
Water’s probe radius is set to 1.40 to reproduce solvation energies properly. All other probe radii are calculated from r3 = (3mΔ)/(4πρ) (1024Å3/cm3), where r is the solvent probe radius in angstroms, m is the molecular mass obtained by dividing the molecular weight given in Ref. [104] in g mol−1 by 6.02 x 1023, Δ is the packing density, and ρ is the density in g/cm3 at 20°C obtained from ref. [104]. Finding the actual Δ would require a detailed knowledge of the structure of the liquid. Currently, all Δ values for these liquids are assumed to be 0.5. (For FCC lattices, Δ is 0.7405, and for BCC lattices, Δ is 0.6802.)
If you use water as the solvent, you have the choice of two methods of defining the cavity. The default is to use the van der Waals radii, as defined in the pbf_radii.ark file (or in the old default.lewis file, if pbf_radii=0). The alternative is to use an isosurface of the density: that is, the cavity is the surface at a specified, small value of the electron density. To use this method, set pbf_isodens=1. This method removes the need to define fixed van der Waals radii for each atom type, and automatically adjusts to changes in the electron density (as a real solvent would). It includes an explicit solvent correction factor based on the assumption that the interaction with the solvent is a function of the electric field on the molecular surface [272]. When you use this method, you should include diffuse functions in the basis set, which are important for polar groups and for anions. This method is particularly recommended for computing the solvation energy of anions.
Note: The explicit correction is known to overestimate the interaction with the solvent for systems containing nitro groups or multiple fluorine atoms.
Several keywords are available for the control of the finite-element mesh used by the Poisson-Boltzmann solver. To increase the resolution, you can increase the values of pbf_nshells, pbf_molsurf_idx, pbf_protosph_res, and pbf_boundary_res and decrease the values of pbf_fem_mxres_cg.
|
Keyword |
Value |
Description |
|
0 |
Do not include solute cavity energy term in solvation calculation |
|
|
|
1 |
Include solute cavity energy term for water, do not include it for other solvents |
|
|
2 |
Force calculation of cavity energy term for all solvents |
|
0 |
Do not include the first shell correction factor term in the solvation energy |
|
|
|
1 |
Include the first shell correction factor term in the solvation energy (default for most calculations in water; turns on ivanset=1 by default). Ignored if pbf_isodens=1 or icavity = 0 or the solvent is not water. |
|
|
2 |
Force calculation of first shell correction factor term for all solvents. Ignored if icavity = 0 or if the solvent is not water and icavity=1. |
|
0 |
Do not set van der Waals radii according to Lewis structure |
|
|
|
1 |
Set van der Waals radii according to the Lewis structure specified by lewstr (1 by default); see LMP2 Keywords in the Jaguar Input File. Only used if pbf_isodens=0 and pbf_radii=0. |
|
0 |
Combine terms for all one-electron matrices |
|
|
|
1 |
Keep kinetic energy terms, nuclear attraction integrals, and point charge terms separate (Note: if isolv=2, kesep=1 by default) |
|
0 |
Compute gradients in solvent with method used in Jaguar 3.5 and earlier. |
|
|
|
1 |
Compute gradients in solvent with robust Jaguar 4.0 method. |
|
|
2 |
Compute gradients in solvent with improved Jaguar 7.8 method. |
|
9 |
Number of shells around each atom in the finite-element mesh. |
|
|
4 |
The index of the shell to use for the molecular surface. |
|
|
21 |
The Lebedev angular scheme (see Table 1) on the prototype atomic sphere. |
|
|
14 |
The Lebedev angular scheme (see Table 1) at the dielectric boundary. |
|
|
10−4 |
The maximum residual for the charge scaling solver. |
|
|
0 |
Do not use the density isosurface to construct the PBF grid. |
|
|
|
1 |
Use the density isosurface to construct the PBF grid, rather than using preset van der Waals radii. This method is not available with ECPs, and can only be used for water. If you set this keyword, pbf_radii, ivanset, and isurf are ignored. |
|
0.001 |
Value of the density for the isosurface. |
|
|
10−50 |
Value of the isodensity on the molecular boundary. |
|
|
0 |
Use the |
|
|
|
1 |
Use the |
|
|
2 |
Use the |
| 4 | Use the LAD-PBF method to define the van der Waals radii. The radii are adjusted for each molecule in a scheme similar to that described in the work of Plett et al. | |
|
8.0 |
Factor controlling the resolution of the inner surfaces. |
|
|
16.0 |
Factor controlling the resolution of the outer surfaces, excluding boundary. |
|
|
0.2 |
Resolution on the boundary shell. |
|
|
2.2 |
Resolution on the molecular surface. |
|
|
30 |
Maximum number of attempts to reduce an inner density that is too large. |
|
|
1.0 |
Inner dielectric constant of solvent [252] |
|
|
1.40 |
Radius of solvent probe molecule |
|
|
100 |
Maximum number of PBF solvation iterations. |
SM8 model
Additional solvents for SM8 can be specified by means of several descriptors, for which the keywords are listed in Table 3. To use a custom solvent, the dielectric constant, index of refraction, acidity and basicity can be derived from experiment, or from interpolation or extrapolation of data available for other solvents, and the fraction of aromatic and halogen atoms can be computed from the chemical formula of the solvent. Solvent parameters for common organic solvents are tabulated in the Minnesota Solvent Descriptor Database, available at: http://comp.chem.umn.edu/solvation/mnsddb.pdf.
|
Keyword |
Description |
|
epsout |
Dielectric constant ε |
|
Abraham’s hydrogen bond acidity α, also called Σα2H |
|
|
Abraham’s hydrogen bond acidity β, also called Σβ2H |
|
|
Aromaticity: the fraction of non-hydrogen solvent atoms that are aromatic carbon atoms |
|
|
Macroscopic surface tension at the air/solvent interface at 298 K, in cal mol−1 Å−2 |
|
|
Electronegative halogenicity: fraction of non-hydrogen solvent atoms that are F, Cl, Br |
|
|
index of refraction at optical frequencies at 293 K, n ≡ n20D |
PCM Model
The polarizable continuum models (PCM) use a molecule-shaped cavity and the full molecular electrostatic potential to obtain apparent surface charge. They include the conductor-like screening models, COSMO [232], GCOSMO [233], and C-PCM [234], and the surface simulation of volume polarization for electrostatics (SS(v)PE [235], which is also known as the "integral equation formalism" (IEF-PCM) [236, 237]. The COSMO and C-PCM models are closely related: the only difference between them comes from a dielectric charge screening factor in the form
−(εsolv − 1.0) / (εsolv + s)
where εsolv is the solvent dielectric constant (set by keyword epsout) or the solvent optical dielectric constant, for TDDFT calculations (set by keyword epsout_opt); and s is the dielectric charge screening factor, set by the keyword pcm_dcs, with the value 0.5 in the COSMO model and 0.0 in the C-PCM model.
The cavity surface is a crucial aspect of PCMs, and the computed properties are quite sensitive to the details of the cavity construction. In Jaguar, the PCM cavity surface is discretized using atom-centered Lebedev grids with specified atomic radii, set with the keywords pcm_vdwscale and pcm_radii. The number of grid points distributed on the sphere surface can be set with the keywords pcm_lebedev_hatom and pcm_lebedev_heavy. Grid points can be discarded if their switching function is less than a threshold, defined by the pcm_switch_thresh keyword.
The discretization procedure usually generates discontinuous potential energy surfaces when the solute system is updated. This is because the surface grid points may emerge from, or disappear within, the solute cavity as the atomic spheres that define the cavity are moved. This can make geometry optimizations very difficult to converge, and produce large errors in vibrational frequency calculations. Jaguar uses smoothing schemes to overcome the problems of the discontinuous potential energy surface, following either the work of Lange and Herbert [238, 239], or the work of York and Karplus [240]. The smoothing scheme can be selected with the keyword pcm_smooth_method.
For the COSMO model, Jaguar can store the screening charge surface in a .cosmo file that can be used as input for the COSMOTherm program. Set iwrt_cosmo=1 to write a COSMOTherm input file. COSMOTherm is not part of the software installation and is not distributed by Schrödinger.
|
Keyword |
Value |
Description |
|---|---|---|
|
real |
Solvent optical dielectric constant, used for TDDFT calculations (square of refractive index) |
|
|
|
Use the COSMO model [232]. |
|
|
|
|
Use the C-PCM model [234]. |
|
|
|
Use the SS(v)PE model [235]. |
|
real |
Dielectric charge screening factor. The value is 0.5 in the COSMO model, 0.0 in the C-PCM model. |
|
|
1.2 |
Van der Waals radius scaling factor, used to define the cavity surface. |
|
|
|
Use Bondi (van der Waals) atomic radii [241], multiplied by the van der Waals radius scaling factor, pcm_vdwscale. |
|
|
|
|
Use Universal Force Field (UFF) atomic radii [105]. |
|
|
|
Use Klamt atomic radii [242]. |
|
|
|
Use the DRACO scaled Bondi atomic radii [328]. Only available for solvent_model=pcm (pcm_model= cpcm or cosmo) and solvent_model=smd (pcm_model=cpcm). |
|
86 |
Number of Lebedev grid points for H atoms. |
|
|
146 |
Number of Lebedev grid points for non-H (heavy) atoms. |
|
|
10−8 |
Threshold for discarding grid points based on their switching function. |
|
|
|
||
|
|
|
Use the smoothing method of York and Karplus [240]. |
| iwrt_cosmo | 0 | Do not write a COSMOTherm input file |
| 1 | Write a COSMOTherm input file |
Solvation entropy keywords
To calculate solvation entropy with a custom solvent, set the keyword solvent=other in the gen section of the Jaguar input file. You must then supply the following parameters for the solvent:
|
Keyword |
Description |
|
Pitzer acentric factor ω. Needed for Sω. Available by default for some solvents. |
|
|
Dielectric constant ε. Needed for Sϵ and Sϵα. Available by default for all defined solvents. |
|
|
Thermal expansion coefficient α. Needed for Sϵα. Available by default for some solvents. |
|
|
Molecular weight, in g/mol. |
|
|
Density, in g/mL. |
|
|
Volume of a single solvent molecule, in ang3. |
|
|
Surface area of a single solvent molecule, in ang2. |
|
|
Radius of gyration of a single solvent molecule, in ang. |