lammps/doc/pair_eam.html

449 lines
19 KiB
HTML
Raw Normal View History

<HTML>
<CENTER><A HREF = "http://lammps.sandia.gov">LAMMPS WWW Site</A> - <A HREF = "Manual.html">LAMMPS Documentation</A> - <A HREF = "Section_commands.html#comm">LAMMPS Commands</A>
</CENTER>
<HR>
<H3>pair_style eam command
</H3>
<H3>pair_style eam/opt command
</H3>
<H3>pair_style eam/alloy command
</H3>
<H3>pair_style eam/alloy/opt command
</H3>
<H3>pair_style eam/cd command
</H3>
<H3>pair_style eam/fs command
</H3>
<H3>pair_style eam/fs/opt command
</H3>
<P><B>Syntax:</B>
</P>
<PRE>pair_style style
</PRE>
<UL><LI>style = <I>eam</I> or <I>eam/alloy</I> or <I>eam/cd</I> or <I>eam/fs</I> or <I>eam/opt</I> or <I>eam/alloy/opt</I> or <I>eam/fs/opt</I>
</UL>
<P><B>Examples:</B>
</P>
<PRE>pair_style eam
pair_style eam/opt
pair_coeff * * cuu3
pair_coeff 1*3 1*3 niu3.eam
</PRE>
<PRE>pair_style eam/alloy
pair_style eam/alloy/opt
pair_coeff * * ../potentials/nialhjea.eam.alloy Ni Al Ni Ni
</PRE>
<PRE>pair_style eam/cd
pair_coeff * * ../potentials/FeCr.cdeam Fe Cr
</PRE>
<PRE>pair_style eam/fs
pair_style eam/fs/opt
pair_coeff * * nialhjea.eam.fs Ni Al Ni Ni
</PRE>
<P><B>Description:</B>
</P>
<P>Style <I>eam</I> computes pairwise interactions for metals and metal alloys
using embedded-atom method (EAM) potentials <A HREF = "#Daw">(Daw)</A>. The total
energy Ei of an atom I is given by
</P>
<CENTER><IMG SRC = "Eqs/pair_eam.jpg">
</CENTER>
<P>where F is the embedding energy which is a function of the atomic
electron density rho, phi is a pair potential interaction, and alpha
and beta are the element types of atoms I and J. The multi-body
nature of the EAM potential is a result of the embedding energy term.
Both summations in the formula are over all neighbors J of atom I
within the cutoff distance.
</P>
<P>Style <I>eam/opt</I> is an optimized version of style <I>eam</I> that should
give identical answers. Depending on system size and the processor
you are running on, it may be 5-25% faster (for the pairwise portion
of the run time).
</P>
<P>The cutoff distance and the tabulated values of the functionals F,
rho, and phi are listed in one or more files which are specified by
the <A HREF = "pair_coeff.html">pair_coeff</A> command. These are ASCII text files
in a DYNAMO-style format which is described below. DYNAMO was the
original serial EAM MD code, written by the EAM originators. Several
DYNAMO potential files for different metals are included in the
"potentials" directory of the LAMMPS distribution. All of these files
are parameterized in terms of LAMMPS <A HREF = "units.html">metal units</A>.
</P>
<P>IMPORTANT NOTE: The <I>eam</I> style reads single-element EAM potentials in
the DYNAMO <I>funcfl</I> format. Either single element or alloy systems
can be modeled using multiple <I>funcfl</I> files and style <I>eam</I>. For the
alloy case LAMMPS mixes the single-element potentials to produce alloy
potentials, the same way that DYNAMO does. Alternatively, a single
DYNAMO <I>setfl</I> file or Finnis/Sinclair EAM file can be used by LAMMPS
to model alloy systems by invoking the <I>eam/alloy</I> or <I>eam/cd</I> or
<I>eam/fs</I> styles as described below. These files require no mixing
since they specify alloy interactions explicitly.
</P>
<P>Note that unlike for other potentials, cutoffs for EAM potentials are
not set in the pair_style or pair_coeff command; they are specified in
the EAM potential files themselves. Likewise, the EAM potential files
list atomic masses; thus you do not need to use the <A HREF = "mass.html">mass</A>
command to specify them.
</P>
<P>There are several WWW sites that distribute and document EAM
potentials stored in DYNAMO or other formats:
</P>
<PRE>http://www.ctcms.nist.gov/potentials
http://cst-www.nrl.navy.mil/ccm6/ap
http://enpub.fulton.asu.edu/cms/potentials/main/main.htm
</PRE>
<P>These potentials should be usable with LAMMPS, though the alternate
formats would need to be converted to the DYNAMO format used by LAMMPS
and described on this page. The NIST site is maintained by Chandler
Becker (cbecker at nist.gov) who is good resource for info on
interatomic potentials and file formats.
</P>
<HR>
<P>For style <I>eam</I>, potential values are read from a file that is in the
DYNAMO single-element <I>funcfl</I> format. If the DYNAMO file was created
by a Fortran program, it cannot have "D" values in it for exponents.
C only recognizes "e" or "E" for scientific notation.
</P>
<P>Note that unlike for other potentials, cutoffs for EAM potentials are
not set in the pair_style or pair_coeff command; they are specified in
the EAM potential files themselves.
</P>
<P>For style <I>eam</I> a potential file must be assigned to each I,I pair of
atom types by using one or more pair_coeff commands, each with a
single argument:
</P>
<UL><LI>filename
</UL>
<P>Thus the following command
</P>
<PRE>pair_coeff *2 1*2 cuu3.eam
</PRE>
<P>will read the cuu3 potential file and use the tabulated Cu values for
F, phi, rho that it contains for type pairs 1,1 and 2,2 (type pairs
1,2 and 2,1 are ignored). In effect, this makes atom types 1 and 2 in
LAMMPS be Cu atoms. Different single-element files can be assigned to
different atom types to model an alloy system. The mixing to create
alloy potentials for type pairs with I != J is done automatically the
same way that the serial DYNAMO code originally did it; you do not
need to specify coefficients for these type pairs.
</P>
<P><I>Funcfl</I> files in the <I>potentials</I> directory of the LAMMPS
distribution have an ".eam" suffix. A DYNAMO single-element <I>funcfl</I>
file is formatted as follows:
</P>
<UL><LI>line 1: comment (ignored)
<LI>line 2: atomic number, mass, lattice constant, lattice type (e.g. FCC)
<LI>line 3: Nrho, drho, Nr, dr, cutoff
</UL>
<P>On line 2, all values but the mass are ignored by LAMMPS. The mass is
in mass <A HREF = "units.html">units</A>, e.g. mass number or grams/mole for metal
units. The cubic lattice constant is in Angstroms. On line 3, Nrho
and Nr are the number of tabulated values in the subsequent arrays,
drho and dr are the spacing in density and distance space for the
values in those arrays, and the specified cutoff becomes the pairwise
cutoff used by LAMMPS for the potential. The units of dr are
Angstroms; I'm not sure of the units for drho - some measure of
electron density.
</P>
<P>Following the three header lines are three arrays of tabulated values:
</P>
<UL><LI>embedding function F(rho) (Nrho values)
<LI>effective charge function Z(r) (Nr values)
<LI>density function rho(r) (Nr values)
</UL>
<P>The values for each array can be listed as multiple values per line,
so long as each array starts on a new line. For example, the
individual Z(r) values are for r = 0,dr,2*dr, ... (Nr-1)*dr.
</P>
<P>The units for the embedding function F are eV. The units for the
density function rho are the same as for drho (see above, electron
density). The units for the effective charge Z are "atomic charge" or
sqrt(Hartree * Bohr-radii). For two interacting atoms i,j this is used
by LAMMPS to compute the pair potential term in the EAM energy
expression as r*phi, in units of eV-Angstroms, via the formula
</P>
<PRE>r*phi = 27.2 * 0.529 * Zi * Zj
</PRE>
<P>where 1 Hartree = 27.2 eV and 1 Bohr = 0.529 Angstroms.
</P>
<HR>
<P>Style <I>eam/alloy</I> computes pairwise interactions using the same
formula as style <I>eam</I>. However the associated
<A HREF = "pair_coeff.html">pair_coeff</A> command reads a DYNAMO <I>setfl</I> file
instead of a <I>funcfl</I> file. <I>Setfl</I> files can be used to model a
single-element or alloy system. In the alloy case, as explained
above, <I>setfl</I> files contain explicit tabulated values for alloy
interactions. Thus they allow more generality than <I>funcfl</I> files for
modeling alloys.
</P>
<P>Style <I>eam/alloy/opt</I> is an optimized version of style <I>eam/alloy</I>
that should give identical answers. Depending on system size and the
processor you are running on, it may be 5-25% faster (for the pairwise
portion of the run time).
</P>
<P>For style <I>eam/alloy</I>, potential values are read from a file that is
in the DYNAMO multi-element <I>setfl</I> format, except that element names
(Ni, Cu, etc) are added to one of the lines in the file. If the
DYNAMO file was created by a Fortran program, it cannot have "D"
values in it for exponents. C only recognizes "e" or "E" for
scientific notation.
</P>
<P>Only a single pair_coeff command is used with the <I>eam/alloy</I> style
which specifies a DYNAMO <I>setfl</I> file, which contains information for
M elements. These are mapped to LAMMPS atom types by specifying N
additional arguments after the filename in the pair_coeff command,
where N is the number of LAMMPS atom types:
</P>
<UL><LI>filename
<LI>N element names = mapping of <I>setfl</I> elements to atom types
</UL>
<P>As an example, the potentials/nialhjea <I>setfl</I> file has tabulated EAM
values for 3 elements and their alloy interactions: Ni, Al, and H. If
your LAMMPS simulation has 4 atoms types and you want the 1st 3 to be
Ni, and the 4th to be Al, you would use the following pair_coeff
command:
</P>
<PRE>pair_coeff * * nialhjea.eam.alloy Ni Ni Ni Al
</PRE>
<P>The 1st 2 arguments must be * * so as to span all LAMMPS atom types.
The first three Ni arguments map LAMMPS atom types 1,2,3 to the Ni
element in the <I>setfl</I> file. The final Al argument maps LAMMPS atom
type 4 to the Al element in the <I>setfl</I> file. Note that there is no
requirement that your simulation use all the elements specified by the
<I>setfl</I> file.
</P>
<P>If a mapping value is specified as NULL, the mapping is not performed.
This can be used when an <I>eam/alloy</I> potential is used as part of the
<I>hybrid</I> pair style. The NULL values are placeholders for atom types
that will be used with other potentials.
</P>
<P><I>Setfl</I> files in the <I>potentials</I> directory of the LAMMPS distribution
have an ".eam.alloy" suffix. A DYNAMO multi-element <I>setfl</I> file is
formatted as follows:
</P>
<UL><LI>lines 1,2,3 = comments (ignored)
<LI>line 4: Nelements Element1 Element2 ... ElementN
<LI>line 5: Nrho, drho, Nr, dr, cutoff
</UL>
<P>In a DYNAMO <I>setfl</I> file, line 4 only lists Nelements = the # of
elements in the <I>setfl</I> file. For LAMMPS, the element name (Ni, Cu,
etc) of each element must be added to the line, in the order the
elements appear in the file.
</P>
<P>The meaning and units of the values in line 5 is the same as for the
<I>funcfl</I> file described above. Note that the cutoff (in Angstroms) is
a global value, valid for all pairwise interactions for all element
pairings.
</P>
<P>Following the 5 header lines are Nelements sections, one for each
element, each with the following format:
</P>
<UL><LI>line 1 = atomic number, mass, lattice constant, lattice type (e.g. FCC)
<LI>embedding function F(rho) (Nrho values)
<LI>density function rho(r) (Nr values)
</UL>
<P>As with the <I>funcfl</I> files, only the mass (in mass <A HREF = "units.html">units</A>,
e.g. mass number or grams/mole for metal units) is used by LAMMPS from
the 1st line. The cubic lattice constant is in Angstroms. The F and
rho arrays are unique to a single element and have the same format and
units as in a <I>funcfl</I> file.
</P>
<P>Following the Nelements sections, Nr values for each pair potential
phi(r) array are listed for all i,j element pairs in the same format
as other arrays. Since these interactions are symmetric (i,j = j,i)
only phi arrays with i >= j are listed, in the following order: i,j =
(1,1), (2,1), (2,2), (3,1), (3,2), (3,3), (4,1), ..., (Nelements,
Nelements). Unlike the effective charge array Z(r) in <I>funcfl</I> files,
the tabulated values for each phi function are listed in <I>setfl</I> files
directly as r*phi (in units of eV-Angstroms), since they are for atom
pairs.
</P>
<HR>
<P>Style <I>eam/cd</I> is similar to the <I>eam/alloy</I> style, except that it
computes alloy pairwise interactions using the concentration-dependent
embedded-atom method (CD-EAM). This model can reproduce the enthalpy
of mixing of alloys over the full composition range, as described in
<A HREF = "#Stukowski">(Stukowski)</A>.
</P>
<P>The pair_coeff command is specified the same as for the <I>eam/alloy</I>
style. However the DYNAMO <I>setfl</I> file must has two
lines added to it, at the end of the file:
</P>
<UL><LI>line 1: Comment line (ignored)
<LI>line 2: N Coefficient0 Coefficient1 ... CoeffincientN
</UL>
<P>The last line begins with the degree <I>N</I> of the polynomial function
<I>h(x)</I> that modifies the cross interaction between A and B elements.
Then <I>N+1</I> coefficients for the terms of the polynomial are then
listed.
</P>
<P>Modified EAM <I>setfl</I> files used with the <I>eam/cd</I> style must contain
exactly two elements, i.e. in the current implementation the <I>eam/cd</I>
style only supports binary alloys. The first and second elements in
the input EAM file are always taken as the <I>A</I> and <I>B</I> species.
</P>
<P><I>CD-EAM</I> files in the <I>potentials</I> directory of the LAMMPS
distribution have a ".cdeam" suffix.
</P>
<HR>
<P>Style <I>eam/fs</I> computes pairwise interactions for metals and metal
alloys using a generalized form of EAM potentials due to Finnis and
Sinclair <A HREF = "#Finnis">(Finnis)</A>. The total energy Ei of an atom I is
given by
</P>
<CENTER><IMG SRC = "Eqs/pair_eam_fs.jpg">
</CENTER>
<P>This has the same form as the EAM formula above, except that rho is
now a functional specific to the atomic types of both atoms I and J,
so that different elements can contribute differently to the total
electron density at an atomic site depending on the identity of the
element at that atomic site.
</P>
<P>Style <I>eam/fs/opt</I> is an optimized version of style <I>eam/fs</I> that
should give identical answers. Depending on system size and the
processor you are running on, it may be 5-25% faster (for the pairwise
portion of the run time).
</P>
<P>The associated <A HREF = "pair_coeff.html">pair_coeff</A> command for style <I>eam/fs</I>
reads a DYNAMO <I>setfl</I> file that has been extended to include
additional rho_alpha_beta arrays of tabulated values. A discussion of
how FS EAM differs from conventional EAM alloy potentials is given in
<A HREF = "#Ackland1">(Ackland1)</A>. An example of such a potential is the same
author's Fe-P FS potential <A HREF = "#Ackland2">(Ackland2)</A>. Note that while FS
potentials always specify the embedding energy with a square root
dependence on the total density, the implementation in LAMMPS does not
require that; the user can tabulate any functional form desired in the
FS potential files.
</P>
<P>For style <I>eam/fs</I>, the form of the pair_coeff command is exactly the
same as for style <I>eam/alloy</I>, e.g.
</P>
<PRE>pair_coeff * * nialhjea.eam.fs Ni Ni Ni Al
</PRE>
<P>where there are N additional arguments after the filename, where N is
the number of LAMMPS atom types. The N values determine the mapping
of LAMMPS atom types to EAM elements in the file, as described above
for style <I>eam/alloy</I>. As with <I>eam/alloy</I>, if a mapping value is
NULL, the mapping is not performed. This can be used when an <I>eam/fs</I>
potential is used as part of the <I>hybrid</I> pair style. The NULL values
are used as placeholders for atom types that will be used with other
potentials.
</P>
<P>FS EAM files include more information than the DYNAMO <I>setfl</I> format
files read by <I>eam/alloy</I>, in that i,j density functionals for all
pairs of elements are included as needed by the Finnis/Sinclair
formulation of the EAM.
</P>
<P>FS EAM files in the <I>potentials</I> directory of the LAMMPS distribution
have an ".eam.fs" suffix. They are formatted as follows:
</P>
<UL><LI>lines 1,2,3 = comments (ignored)
<LI>line 4: Nelements Element1 Element2 ... ElementN
<LI>line 5: Nrho, drho, Nr, dr, cutoff
</UL>
<P>The 5-line header section is identical to an EAM <I>setfl</I> file.
</P>
<P>Following the header are Nelements sections, one for each element I,
each with the following format:
</P>
<UL><LI>line 1 = atomic number, mass, lattice constant, lattice type (e.g. FCC)
<LI>embedding function F(rho) (Nrho values)
<LI>density function rho(r) for element I at element 1 (Nr values)
<LI>density function rho(r) for element I at element 2
<LI>...
<LI>density function rho(r) for element I at element Nelement
</UL>
<P>The units of these quantities in line 1 are the same as for <I>setfl</I>
files. Note that the rho(r) arrays in Finnis/Sinclair can be
asymmetric (i,j != j,i) so there are Nelements^2 of them listed in the
file.
</P>
<P>Following the Nelements sections, Nr values for each pair potential
phi(r) array are listed in the same manner (r*phi, units of
eV-Angstroms) as in EAM <I>setfl</I> files. Note that in Finnis/Sinclair,
the phi(r) arrays are still symmetric, so only phi arrays for i >= j
are listed.
</P>
<HR>
<P><B>Mixing, shift, table, tail correction, restart, rRESPA info</B>:
</P>
<P>For atom type pairs I,J and I != J, where types I and J correspond to
two different element types, mixing is performed by LAMMPS as
described above with the individual styles. You never need to specify
a pair_coeff command with I != J arguments for the eam styles.
</P>
<P>This pair style does not support the <A HREF = "pair_modify.html">pair_modify</A>
shift, table, and tail options.
</P>
<P>The eam pair styles do not write their information to <A HREF = "restart.html">binary restart
files</A>, since it is stored in tabulated potential files.
Thus, you need to re-specify the pair_style and pair_coeff commands in
an input script that reads a restart file.
</P>
<P>The eam pair styles can only be used via the <I>pair</I> keyword of the
<A HREF = "run_style.html">run_style respa</A> command. They do not support the
<I>inner</I>, <I>middle</I>, <I>outer</I> keywords.
</P>
<HR>
<P><B>Restrictions:</B>
</P>
<P>All of these styles except those ending in <I>opt</I> and the <I>eam/cd</I>
style are part of the "manybody" package. They are only enabled if
LAMMPS was built with that package (which it is by default).
</P>
<P>The styles ending in <I>opt</I> are part of the "opt" package and also
require the "manybody" package. They are only enabled if LAMMPS was
built with those packages. See the <A HREF = "Section_start.html#2_3">Making
LAMMPS</A> section for more info.
</P>
<P>The <I>eam/cd</I> style is part of the "user-cd-eam" package and also
requires the "manybody" package. It is only enabled if LAMMPS was
built with those packages. See the <A HREF = "Section_start.html#2_3">Making
LAMMPS</A> section for more info.
</P>
<P><B>Related commands:</B>
</P>
<P><A HREF = "pair_coeff.html">pair_coeff</A>
</P>
<P><B>Default:</B> none
</P>
<HR>
<A NAME = "Ackland1"></A>
<P><B>(Ackland1)</B> Ackland, Condensed Matter (2005).
</P>
<A NAME = "Ackland2"></A>
<P><B>(Ackland2)</B> Ackland, Mendelev, Srolovitz, Han and Barashev, Journal
of Physics: Condensed Matter, 16, S2629 (2004).
</P>
<A NAME = "Daw"></A>
<P><B>(Daw)</B> Daw, Baskes, Phys Rev Lett, 50, 1285 (1983).
Daw, Baskes, Phys Rev B, 29, 6443 (1984).
</P>
<A NAME = "Finnis"></A>
<P><B>(Finnis)</B> Finnis, Sinclair, Philosophical Magazine A, 50, 45 (1984).
</P>
<A NAME = "Stukowski"></A>
<P><B>(Stukowski)</B> Stukowski, Sadigh, Erhart, Caro; Modeling Simulation
Materials Science & Engineering, 7, 075005 (2009).
</P>
</HTML>