Overview
This section covers the QMFF-VIII force field as it is implemented into the towhee_ff_QMFF-VIII file in the ForceFields directory.
This force field was designed for a variety of organic liquids.
Note that this is a Class-2 9-6 force field that uses the 'Sixth Power' mixing rules.
Any discrepencies (especially typos) from the published values are the sole responsibility of Marcus Martin,
and I welcome feedback on how this implementation compares with other programs.
References for QMFF-VIII
This forcefield is described in a single paper, and the supplementary information for that paper.
Typos and comments for QMFF-VIII
- The supplementary information contains a few additional parameters that were not mentioned in the main paper.
These have been implemented, although I am not positive about how they were intended to be used.
- The supplementary information does not contain the c= h* and n= parameters. I believe that a formatting error
has replaced those symbols with additional c h and n parameters because there are duplicates of those listed there.
I have been unable to get a useful response from any of the authors and therefore have used some logic combined
with inspiration in order to attempt to reproduce the true values from the duplicates. Here I list all of the
changes I made to the supplementary information in order to create the values implemented into Towhee.
- The h and h* atom types are distinct, but given that they are mixed together due to a
formatting error in the file, and that they are bonded to different atom types, all of their bonded terms
are treated as a generic h. This will work properly so long as the h atom types are only
bonded to carbon and sulfur while the h* atom types are only bonded to oxygen and nitrogen.
- The c= term was erroneously listed as an additional c term in the supplementary information.
Any time a duplicate term is listed that includes a c I assume that this was actually supposed to
be a c=. Given that most of the terms are listed in order in that file it is relatively easy to
figure out which terms should be c=.
- The n= term was erroneously listed as an additional n term in the supplementary information.
Any time a duplicate term is listed that includes a n I assume that this was actually supposed to
be a n=. Given that most of the terms are listed in order in that file it is relatively easy to
figure out which terms should be n=.
- An additional oxygen term is listed and I do not know the purpose for which it was originally intended.
I have labeled the second of any duplicate oxygens as o?.
- Page 50 Line 29: assumed that the duplicate van der Waals listing for c was intended for c=
- Page 50 Line 38: assumed that the duplicate van der Waals listing for h was intended for h*
- Page 50 Line 44: assumed that the duplicate van der Waals listing for n was intedned for n=
- Page 51 Line 5: a duplicate van der Waals listing for o is on this line is a mystery and was set
to the mystery oxygen o?
QMFF-VIII in Towhee
The official force field name for QMFF-VIII in Towhee is 'QMFF-VIII'.
Here is the list of atom names for use in the towhee_input file, along with a brief description.
These names are from the main text of the paper, and also include atoms from the supplementary information that are lacking a description.
Please note that the capitalization and spacing pattern is important and must be followed exactly as listed here.
- Bromine
- 'Br' : no explanation provided. This was assigned a mass appropriate for bromine and
I suspect it is intended for Br- ion.
- 'br' : no explanation provided. This was assigned a mass appropriate for bromine and
I suspect it is intended for a bonded bromine, but while bond increment parameters are included,
there are no bond vibration parameters so it is not usable as currently implemented.
- Calcium
- 'Ca' : no explanation provided. This was assigned a mass appropriate for calcium and
I suspect it is intended for Ca2+ ion.
- Carbon
- 'c' : carbon (sp3) in alkanes
- 'c"' : carbonyl carbon except in amides. Be sure that you are surrounding this atom type with
single quotes in towhee_input as using double quotes will cause a read error for c"
- 'c`' : carbonyl carbon in amides (connected to nitrogen). This is listed as c' in their
work, but is implemented using the alternate style of single quote in order to avoid problems reading
character strings surrounded by the typical Fortran single quotes.
- 'c+' : triply connected carbon in amidine and imidazole cations.
- 'c-' : carbon in carboxylate groups
- 'c=' : carbon (sp2) in alkenes. This atom type was erronously listed as c in
the supplementary information so some inference was required in order to determine which parameters
belong to c= instead of c. See the Typos and Comments section above for a complete
list of inferred parameters for this atom type.
- 'co' : anomeric carbon in acetals and hemiacetals.
- 'cp' : carbon in aromatic rings.
- 'cpb' : bridging carbon in bridged aromatics
- 'cr' : no explanation provided. This was assigned a mass appropriate for carbon and I suspect
it is some form of bonded carbon. The only other parameters provided for this type are bond increment
values with a few forms of nitrogen so this is probably not useful as implemented into Towhee.
- 'ct' : carbon (sp) in alkynes.
- Chlorine
- 'Cl' : no explanation provided. This was assigned a mass appropriate for chlorine and
I suspect it is intended for Cl- ion.
- 'cl' : chlorine. This appears to be intended for use when bonded to other atoms.
- Cesium
- 'Cs' : no explanation provided. This was assigned a mass appropriate for cesium and
I suspect it is intended for Cs+ ion.
- Copper
- 'Cu' : no explanation provided. This was assigned a mass appropriate for copper and
I suspect it is intended for a Cu ion, but I am not sure of the charge state.
- Fluorine
- 'F' : no explanation provided. This was assigned a mass appropriate for fluorine and
I suspect it is intended for F- ion.
- 'f' : fluorine. This appears to be intended for use when bonded to other atoms.
- Hydrogen
- 'h' : hydrogen connected to carbon or sulfur.
- 'h*' : hydrogen connected to oxygen or nitrogen (except in ammonium). This atom type was
erroneously listed as h in the supplementary information so some inference was required in order
to determine which parameters belong to h* insteda of h. See the Typos and Comments
section above for a complete list of inferred parameters for this atom type.
- 'h+' : hydrogen in ammonium groups.
- Iron
- 'Fe' : no explanation provided. This was assigned a mass appropriate for iron and
I suspect it is intended for an Fe ion, but I am not sure of the charge state.
- Iodine
- 'I' : no explanation provided. This was assigned a mass appropriate for iodine and
I suspect it is intended for I- ion.
- 'i' : no explanation provided. This was assigned a mass appropriate for iodine and I suspect
it is intended for iodine bonded to another atom. Unforunately the only other parameters provided for
this atom type are bond increments so this is probably not useful as implemented into Towhee.
- Potassium
- 'K' : no explanation provided. This was assigned a mass appropriate for potassium and
I suspect it is intended for K+ ion.
- Lithium
- 'Li' : no explanation provided. This was assigned a mass appropriate for lithium and
I suspect it is intended for Li+ ion.
- Magnesium
- 'Mg' : no explanation provided. This was assigned a mass appropriate for magnesium and
I suspect it is intended for Mg2+ ion.
- Nitrogen
- 'n' : nitrogen in amides and amidines
- 'n`' : nitrogen in nitro groups. This is listed as n' in their
work, but is implemented using the alternate style of single quote in order to avoid problems reading
character strings surrounded by the typical Fortran single quotes.
- 'n+' : nitrogen in ammonium groups.
- 'n=' : doubly bonded nitrogen. This atom type was erronously listed as n in
the supplementary information so some inference was required in order to determine which parameters
belong to n= instead of n. See the Typos and Comments section above for a complete
list of inferred parameters for this atom type.
- 'na' : nitrogen (sp3) in amines
- 'nh' : nitrogen (sp2) triply connected in five- and six-membered rings (e.g. pyrrole)
- 'nn' : nitrogen (sp2) connected to aromatic rings (e.g. aniline)
- 'np' : nitrogen (sp2) doubly connected in five- and six-membered rings (e.g. pyridine)
- Oxygen
- 'o' : doubly connected oxygen
- 'o`' : carbonyl and sulfonyl oxygen. This is listed as o' in their
work, but is implemented using the alternate style of single quote in order to avoid problems reading
character strings surrounded by the typical Fortran single quotes.
- 'o-' : oxygen in carboxylates, dialkyl phosphaes, and phosphate dianions
- 'op' : oxygen in aromatic rings
- 'o?' : no explanation provided. There are several instances of duplicate oxygen parameters and
I suspect it is an additional unpublished oxygen value that got mixed up with the normal o in
the formatting problems. I have given it the o? name and included it in case anyone knows
its true purpose.
- Phosphorus
- 'p' : phosphorus in dialkyl phosphates
- 'p-' : phosphorus in phosphate dianions
- Sodium
- 'Na' : no explanation provided. This was assigned a mass appropriate for sodium and
I suspect it is intended for Na+ ion.
- Sulfur
- 's' : sulfur (sp3) doubly connected
- 's"' : sulfur in solfones and solfonamides. Be sure that you are surrounding this atom type with
single quotes in towhee_input as using double quotes will cause a read error for s"
- 'sp' : sulfur in aromatic rings
- Rubidium
- 'Rb' : no explanation provided. This was assigned a mass appropriate for rubidium and
I suspect it is intended for Mg+ ion.
- Zinc
- 'Zn' : no explanation provided. This was assigned a mass appropriate for zinc and
I suspect it is intended for a Zn ion, but am not sure of the charge state.
Coulombic interactions
This force field uses point charges and has been set up to assign the point charges using the bond increment method.
Return to the Towhee Capabilities web page
|