1
"LAMMPS WWW Site"_lws - "LAMMPS Documentation"_ld - "LAMMPS Commands"_lc :c
3
:link(lws,http://lammps.sandia.gov)
5
:link(lc,Section_commands.html#comm)
9
fix qeq/point command :h3
10
fix qeq/shielded command :h3
11
fix qeq/slater command :h3
12
fix qeq/dynamic command :h3
16
fix ID group-ID style Nevery cutoff tolerance maxiter qfile :pre
18
ID, group-ID are documented in "fix"_fix.html command
19
style = {qeq/point} or {qeq/shielded} or {qeq/slater} or {qeq/dynamic}
20
Nevery = perform charge equilibration every this many steps
21
cutoff = global cutoff for charge-charge interactions (distance unit)
22
tolerance = precision to which charges will be equilibrated
23
maxiter = maximum iterations to perform charge equilibration
24
qfile = a filename with QEq parameters :ul
28
fix 1 all qeq/point 1 10 1.0e-6 200 param.qeq1
29
fix 1 qeq qeq/shielded 1 8 1.0e-6 100 param.qeq2
30
fix 1 all qeq/slater 5 10 1.0e-6 100 params
31
fix 1 qeq qeq/dynamic 1 12 1.0e-3 100 my_qeq :pre
35
Perform the charge equilibration (QEq) method as described in "(Rappe
36
and Goddard)"_#Rappe and formulated in "(Nakano)"_#Nakano (also known
37
as the matrix inversion method) and in "(Rick and Stuart)"_#Rick (also
38
known as the extended Lagrangian method) based on the
39
electronegativity equilization principle.
41
These fixes can be used with any "pair style"_pair_style.html in
42
LAMMPS, so long as per-atom charges are defined. The most typical
43
use-case is in conjunction with a "pair style"_pair_style.html that
44
performs charge equilibration periodically (e.g. every timestep), such
45
as the ReaxFF or Streitz-Mintmire potential (the latter is not yet
46
implemented in LAMMPS). But these fixes can also be used with
47
potentials that normally assume per-atom charges are fixed, e.g. a
48
"Buckingham"_pair_buck.html or "LJ/Coulombic"_pair_lj.html potential.
50
Because the charge equilibration calculation is effectively
51
independent of the pair style, these fixes can also be used to perform
52
a one-time assignment of charges to atoms. For example, you could
53
define the QEq fix, perform a zero-timestep run via the "run"_run.html
54
command without any pair style defined which would set per-atom
55
charges (based on the current atom configuration), then remove the fix
56
via the "unfix"_unfix.html command before performing further dynamics.
58
IMPORTANT NOTE: Computing and using charge values different from
59
published values defined for a fixed-charge potential like Buckingham
60
or CHARMM or AMBER, can have a strong effect on energies and forces,
61
and produces a different model than the published versions.
63
IMPORTANT NOTE: The "fix qeq/comb"_fix_qeq_comb.html command must
64
still be used to perform charge equliibration with the "COMB
65
potential"_pair_comb.html. The "fix qeq/reax"_fix_qeq_reax.html
66
command can be used to perform charge equilibration with the "ReaxFF
67
force field"_pair_reax_c.html, although fix qeq/shielded yields the
68
same results as fix qeq/reax if {Nevery}, {cutoff}, and {tolerance} are
69
the same. Eventually the fix qeq/reax command will be deprecated.
71
The QEq method minimizes the electrostatic energy of the system (or
72
equalizes the derivative of energy with respect to charge of all the
73
atoms) by adjusting the partial charge on individual atoms based on
74
interactions with their neighbors within {cutoff}. It reqires a few
75
parameters, in {metal} units, for each atom type which provided in a
76
file specified by {qfile}. The file has the following format
78
1 chi eta gamma zeta qcore
79
2 chi eta gamma zeta qcore
81
Ntype chi eta gamma zeta qcore :pre
83
There is one line per atom type with the following parameters.
84
Only a subset of the parameters is used by each QEq style as descibed
85
below, thus the others can be set to 0.0 if desired.
87
{chi} = electronegativity in energy units
88
{eta} = self-Coulomb potential in energy units
89
{gamma} = shielded Coulomb constant defined by "ReaxFF force field"_#vanDuin in distance units
90
{zeta} = Slater type orbital exponent defined by the "Streitz-Mintmire"_#Streitz potential in reverse distance units
91
{qcore} = charge of the nucleus defined by the "Streitz-Mintmire potential"_#Streitz potential in charge units :ul
93
The {qeq/point} style describes partial charges on atoms as point
94
charges. Interaction between a pair of charged particles is 1/r,
95
which is the simplest description of the interaction between charges.
96
Only the {chi} and {eta} parameters from the {qfile} file are used.
97
Note that Coulomb catastrophe can occur if repulsion between the pair
98
of charged particles is too weak. This style solves partial charges
99
on atoms via the matrix inversion method. A tolerance of 1.0e-6 is
100
usually a good number.
102
The {qeq/shielded} style describes partial charges on atoms also as
103
point charges, but uses a shielded Coulomb potential to describe the
104
interaction between a pair of charged particles. Interaction through
105
the shielded Coulomb is given by equation (13) of the "ReaxFF force
106
field"_#vanDuin paper. The shielding accounts for charge overlap
107
between charged particles at small separation. This style is the same
108
as "fix qeq/reax"_fix_qeq_reax.html, and can be used with "pair_style
109
reax/c"_pair_reax_c.html. Only the {chi}, {eta}, and {gamma}
110
parameters from the {qfile} file are used. This style solves partial
111
charges on atoms via the matrix inversion method. A tolerance of
112
1.0e-6 is usually a good number.
114
The {qeq/slater} style describes partial charges on atoms as spherical
115
charge densities centered around atoms via the Slater 1{s} orbital, so
116
that the interaction between a pair of charged particles is the
117
product of two Slater 1{s} orbitals. The expression for the Slater
118
1{s} orbital is given under equation (6) of the
119
"Streitz-Mintmire"_#Streitz paper. Only the {chi}, {eta}, {zeta}, and
120
{qcore} parameters from the {qfile} file are used. This style solves
121
partial charges on atoms via the matrix inversion method. A tolerance
122
of 1.0e-6 is usually a good number.
124
The {qeq/dynamic} style describes partial charges on atoms as point
125
charges that interact through 1/r, but the extended Lagrangian method
126
is used to solve partial charges on atoms. Only the {chi} and {eta}
127
parameters from the {qfile} file are used. Note that Coulomb
128
catastrophe can occur if repulsion between the pair of charged
129
particles is too weak. A tolerance of 1.0e-3 is usually a good
132
Note that {qeq/point}, {qeq/shielded}, and {qeq/slater} describe
133
different charge models, whereas the matrix inversion method and the
134
extended Lagrangian method ({qeq/dynamic}) are different solvers.
136
Note that the {qeq/point} and the {qeq/dynamic} styles both describe
137
charges as point charges that interact through 1/r relationship, but
138
solve partial charges on atoms using different solvers. Styles
139
{qeq/point} and the {qeq/dynamic} should yield comparable results if
140
the QEq parameters and {Nevery}, {cutoff}, and {tolerance} are the
141
same. Style {qeq/point} is typically faster, but {qeq/dynamic} scales
142
better on larger sizes.
144
IMPORTANT NOTE: To avoid the evaluation of the derivative of charge
145
with respect to position, which is typically ill-defined, the system
146
should have a zero net charge.
148
IMPORTANT NOTE: Developing QEq parameters (chi, eta, gamma, zeta, and
149
qcore) is an "art". Charges on atoms are not guaranteed to
150
equilibrate with arbitrary choices of these parameters. We do not
151
develop these QEq paramters. See the examples/qeq directory for some
154
[Restart, fix_modify, output, run start/stop, minimize info:]
156
No information about these fixes is written to "binary restart
157
files"_restart.html. No global scalar or vector or per-atom
158
quantities are stored by these fixes for access by various "output
159
commands"_Section_howto.html#howto_15. No parameter of these fixes
160
can be used with the {start/stop} keywords of the "run"_run.html
163
Thexe fixes are invoked during "energy minimization"_minimize.html.
167
These fixes are part of the QEQ package. They are only enabled if
168
LAMMPS was built with that package. See the "Making
169
LAMMPS"_Section_start.html#start_3 section for more info.
173
"fix qeq/reax"_fix_qeq_reax.html, "fix qeq/comb"_fix_qeq_comb.html
180
[(Rappe and Goddard)] A. K. Rappe and W. A. Goddard III, J Physical
181
Chemistry, 95, 3358-3363 (1991).
184
[(Nakano)] A. Nakano, Computer Physics Communications, 104, 59-69 (1997).
187
[(Rick and Stuart)] S. W. Rick, S. J. Stuart, B. J. Berne, J Chemical Physics
191
[(Streitz-Mintmire)] F. H. Streitz, J. W. Mintmire, Physical Review B, 50,
195
[(ReaxFF)] A. C. T. van Duin, S. Dasgupta, F. Lorant, W. A. Goddard III, J
196
Physical Chemistry, 105, 9396-9049 (2001)