News
News

\(polar\)

program to compute molecular polarizabilities, using Thole's empirical model (and adaptations of it)
Current version: 2024.01 (shared with CC-BY-NC license). See also all python tools and software pages.

Atomic element parameters

parameter
description
default
\(atpol(EL)\)
Atomic polarizability for \(EL\), where \(EL\) can be either the element-nr (1 for H, 6 for C, etc.) or the element-name (H, C, etc.)
-
\(ego(i,EL)\)
Atomic gaussian-exponent for \(EL\), where \(EL\) can be either the element-nr (1 for H, 6 for C, etc.) or the element-name (H, C, etc.). There are three \(ego\) exponents per element, as indicated by the index \(i\) (=1,2,3).
-
\(frozen(EL)\) or \(frozen(xFACT)\)
Indicates that \(EL\) will be frozen (i.e. its polarizability and gaussian exponent will not be optimized), where \(EL\) can be either the element-nr (1 for H, 6 for C, etc.) or the element-name (H, C, etc.), or that the parameter \(xFACT\) (i.e. \(afact\), \(bfact\), etc.) are kept frozen in the optimizations.
-
\(gauss(EL)\)
Atomic gaussian-exponent (for ithole=3,4,5) for \(EL\), where \(EL\) can be either the element-nr (1 for H, 6 for C, etc.) or the element-name (H, C, etc.)
-

Parameters

parameter
description
default
\(afact\)
A-factor in Thole's model (and modifications of it)
2.1304
\(alfmin\)
Minimum value for any atomic polarizability
0.2
\(deltacvg\)
Convergence criterium for delta function to be minimized
1.0e-8
\(gssmin\)
Minimum value for any gaussian exponent (with ithole=3,4,5)
0.2
\(idelta\)
Defines the delta function to be minimized when fitting new atomic polarizabilities, based on the computed molecular polarizability \(\alpha_{i}^{comp}\) and reference polarizability \(\alpha_{i}^{ref}\), corresponding X,Y,Z components \(\alpha_{i,A}^{comp}\) or tensor values \(\alpha_{i,PQ}^{comp}\):
  • 1: \(\Delta = \displaystyle\sum_{i}^{N_{mol}} (\alpha_{i}^{comp} - \alpha_{i}^{ref})^{2} \)
  • 2: \(\Delta = \displaystyle\sum_{i}^{N_{mol}} \mid \alpha_{i}^{comp} - \alpha_{i}^{ref} \mid \)
  • 3: \(\Delta = \displaystyle\sum_{i}^{N_{mol}} ( \dfrac{ \alpha_{i}^{comp} - \alpha_{i}^{ref}}{\alpha_{i}^{ref}} )^{2} \)
  • 4: \(\Delta = \displaystyle\sum_{i}^{N_{mol}} \mid \dfrac{\alpha_{i}^{comp} - \alpha_{i}^{ref}}{\alpha_{i}^{ref}} \mid \)
  • 5: \(\Delta = \displaystyle\sum_{i,A}^{N_{mol}} (\alpha_{i,A}^{comp} - \alpha_{i,A}^{ref})^{2} \)
  • 6: \(\Delta = \displaystyle\sum_{i,A}^{N_{mol}} \mid \alpha_{i,A}^{comp} - \alpha_{i,A}^{ref} \mid \)
  • 7: \(\Delta = \displaystyle\sum_{i,A}^{N_{mol}} ( \dfrac{ \alpha_{i,A}^{comp} - \alpha_{i,A}^{ref}}{\alpha_{i,A}^{ref}} )^{2} \)
  • 8: \(\Delta = \displaystyle\sum_{i,A}^{N_{mol}} \mid \dfrac{\alpha_{i,A}^{comp} - \alpha_{i,A}^{ref}}{\alpha_{i,A}^{ref}} \mid \)
  • 9: \(\Delta = \displaystyle\sum_{i,PQ}^{N_{mol}} (\alpha_{i,PQ}^{comp} - \alpha_{i,PQ}^{ref})^{2} \)
1
\(iopt\)
Defines what should be fitted:
  • -1: Check gradient
  • 0: Predict molecular polarizabilities
  • 1: Optimizing atomic polarizabilities
  • 2: Optimizing global parameters (AFACT, etc.)
  • 3: Optimizing GAUSS and/or EGO exponents
  • 4: Optimizing all parameters simultaneously
0
\(iprfrq\)
Printing frequency during optimization of parameters
10
\(iprint\)
Amount of output desired (for debug purposes)
1
\(ithole\)
Type of screening function used in Thole's model:
  • 0: No screening function
  • 1: Conical function
  • 2: Exponential function
  • 3: IM-SQRT function (Jensen et al.)
  • 4: IM-QDRT function (Jensen et al.)
  • 5: IM-ERF function (Jensen et al.)
  • 6-26: experimental options (unpublished)
2
\(iunit\)
Unit used for atomic X,Y,Z coordinates:
  • 0: Coordinates in Bohr
  • 1: Coordinates in Angstrom
0
\(mxiter\)
A-factor in Thole's model (and modifications of it)
2.1304
\(stepmx\)
Maximum step length in optimizing atomic polarizabilities (and/or gaussian exponents in case if ithole=3,4,5)
0.1

Parameters for experimental models

parameter
description
default
\(bfact\)
B-factor in experimental models
0.0
\(cfact\)
C-factor in experimental models
1.0
\(dfact\)
D-factor in experimental models
1.0
\(efact\)
E-factor in experimental models
0.0
\(ffact\)
F-factor in experimental models
0.0
\(gfact\)
G-factor in experimental models
0.0
\(hfact\)
H-factor in experimental models
0.0
\(ifact\)
I-factor in experimental models
0.0
\(jfact\)
J-factor in experimental models
0.0