# pair_style meam command Accelerator Variants: *meam/kk* # pair_style meam/ms command Accelerator Variants: *meam/ms/kk* ## Syntax ``` LAMMPS pair_style style ``` - style = *meam* or *meam/ms* ## Examples ``` LAMMPS pair_style meam pair_coeff * * ../potentials/library.meam Si ../potentials/si.meam Si pair_coeff * * ../potentials/library.meam Ni Al NULL Ni Al Ni Ni pair_style meam/ms pair_coeff * * ../potentials/library.msmeam H Ga ../potentials/HGa.meam H Ga ``` ## Description :::: note ::: title Note ::: The behavior of the MEAM potential for alloy systems has changed as of November 2010; see description below of the mixture_ref_t parameter :::: Pair style *meam* computes non-bonded interactions for a variety of materials using the modified embedded-atom method (MEAM) [(Baskes)](Baskes). Conceptually, it is an extension to the original [EAM method](pair_eam) which adds angular forces. It is thus suitable for modeling metals and alloys with fcc, bcc, hcp and diamond cubic structures, as well as materials with covalent interactions like silicon and carbon. The *meam* pair style is a translation of the original Fortran version to C++. It is functionally equivalent but more efficient and has additional features. The Fortran version of the *meam* pair style has been removed from LAMMPS after the 12 December 2018 release. Pair style *meam/ms* uses the multi-state MEAM (MS-MEAM) method according to [(Baskes2)](Baskes2), which is an extension to MEAM. This pair style is mostly equivalent to *meam* and differs only where noted in the documentation below. In the MEAM formulation, the total energy E of a system of atoms is given by: $$E = \sum_i \left\{ F_i(\bar{\rho}_i) + \frac{1}{2} \sum_{i \neq j} \phi_{ij} (r_{ij}) \right\}$$ where *F* is the embedding energy which is a function of the atomic electron density $\rho$, and $\phi$ is a pair potential interaction. The pair interaction is summed over all neighbors J of atom I within the cutoff distance. As with EAM, the multi-body nature of the MEAM potential is a result of the embedding energy term. Details of the computation of the embedding and pair energies, as implemented in LAMMPS, are given in [(Gullet)](Gullet) and references therein. The various parameters in the MEAM formulas are listed in two files which are specified by the [pair_coeff](pair_coeff) command. These are ASCII text files in a format consistent with other MD codes that implement MEAM potentials, such as the serial DYNAMO code and Warp. Several MEAM potential files with parameters for different materials are included in the \"potentials\" directory of the LAMMPS distribution with a \".meam\" suffix. All of these are parameterized in terms of LAMMPS [metal units](units). Note that unlike for other potentials, cutoffs for MEAM potentials are not set in the pair_style or pair_coeff command; they are specified in the MEAM potential files themselves. Only a single pair_coeff command is used with the *meam* style which specifies two MEAM files and the element(s) to extract information for. The MEAM elements are mapped to LAMMPS atom types by specifying N additional arguments after the second filename in the pair_coeff command, where N is the number of LAMMPS atom types: - MEAM library file - Element1, Element2, \... - MEAM parameter file - N element names = mapping of MEAM elements to atom types See the [pair_coeff](pair_coeff) page for alternate ways to specify the path for the potential files. As an example, the `potentials/library.meam` file has generic MEAM settings for a variety of elements. The `potentials/SiC.meam` file has specific parameter settings for a Si and C alloy system. If your LAMMPS simulation has 4 atoms types and you want the first 3 to be Si, and the fourth to be C, you would use the following pair_coeff command: ``` LAMMPS pair_coeff * * library.meam Si C sic.meam Si Si Si C ``` The first 2 arguments must be \* \* so as to span all LAMMPS atom types. The first filename is the element library file. The list of elements following it extracts lines from the library file and assigns numeric indices to these elements. The second filename is the alloy parameter file, which refers to elements using the numeric indices assigned before. The arguments after the parameter file map LAMMPS atom types to elements, i.e. LAMMPS atom types 1,2,3 to the MEAM Si element. The final C argument maps LAMMPS atom type 4 to the MEAM C element. If the second filename is specified as NULL, no parameter file is read, which simply means the generic parameters in the library file are used. Use of the NULL specification for the parameter file is discouraged for systems with more than a single element type (e.g. alloys), since the parameter file is expected to set element interaction terms that are not captured by the information in the library file. If a mapping value is specified as NULL, the mapping is not performed. This can be used when a *meam* potential is used as part of the *hybrid* pair style. The NULL values are placeholders for atom types that will be used with other potentials. :::: note ::: title Note ::: If the second filename is NULL, the element names between the two filenames can appear in any order, e.g. \"Si C\" or \"C Si\" in the example above. However, if the second filename is **not** NULL (as in the example above), it contains settings that are indexed **by numbers** for the elements that precede it. Thus you need to ensure that you list the elements between the filenames in an order consistent with how the values in the second filename are indexed. See details below on the syntax for settings in the second file. :::: The MEAM library file provided with LAMMPS has the name `potentials/library.meam`. It is the \"meamf\" file used by other MD codes. Aside from blank and comment lines (starting with \# which can appear anywhere), it is formatted as a series of entries, each of which has 19 parameters and can span multiple lines: elt, lat, z, ielement, atwt, alpha, b0, b1, b2, b3, alat, esub, asub, t0, t1, t2, t3, rozero, ibar The *elt* and *lat* parameters are text strings, such as *elt* = Si or Cu and *lat* = dia or fcc. Because the library file is used by Fortran MD codes, these strings may be enclosed in single quotes, but this is not required. The other numeric parameters match values in the formulas above. The value of the *elt* string is what is used in the pair_coeff command to identify which settings from the library file you wish to read in. There can be multiple entries in the library file with the same *elt* value; LAMMPS reads the first matching entry it finds and ignores the rest. Other parameters in the MEAM library file correspond to single-element potential parameters: lat = lattice structure of reference configuration z = number of nearest neighbors in the reference structure ielement = atomic number atwt = atomic weight alat = lattice constant of reference structure esub = energy per atom (eV) in the reference structure at equilibrium asub = "A" parameter for MEAM (see e.g. `(Baskes) `__) The *alpha*, *b0*, *b1*, *b2*, *b3*, *t0*, *t1*, *t2*, *t3* parameters correspond to the standard MEAM parameters in the literature [(Baskes)](Baskes) (the b parameters are the standard beta parameters). Note that only parameters normalized to *t0 = 1.0* are supported. The *rozero* parameter is an element-dependent density scaling that weights the reference background density (see e.g. equation 4.5 in [(Gullet)](Gullet)) and is typically 1.0 for single-element systems. The *ibar* parameter selects the form of the function G(Gamma) used to compute the electron density; options are 0 => G = sqrt(1+Gamma) 1 => G = exp(Gamma/2) 2 => not implemented 3 => G = 2/(1+exp(-Gamma)) 4 => G = sqrt(1+Gamma) -5 => G = +-sqrt(abs(1+Gamma)) If used, the MEAM parameter file contains settings that override or complement the library file settings. Examples of such parameter files are in the potentials directory with a \".meam\" suffix. Their format is the same as is read by other Fortran MD codes. Aside from blank and comment lines (start with \# which can appear anywhere), each line has one of the following forms. Each line can also have a trailing comment (starting with #) which is ignored. keyword = value keyword(I) = value keyword(I,J) = value keyword(I,J,K) = value The indices I, J, K correspond to the elements selected from the MEAM library file numbered in the order of how those elements were selected starting from 1. Thus for the example given before ``` LAMMPS pair_coeff * * library.meam Si C sic.meam Si Si Si C ``` an index of 1 would refer to Si and an index of 2 to C. The recognized keywords for the parameter file are as follows: rc = cutoff radius for cutoff function; default = 4.0 delr = length of smoothing distance for cutoff function; default = 0.1 rho0(I) = relative density for element I (overwrites value read from meamf file) Ec(I,J) = cohesive energy of reference structure for I-J mixture delta(I,J) = heat of formation for I-J alloy; if Ec_IJ is input as zero, then LAMMPS sets Ec_IJ = (Ec_II + Ec_JJ)/2 - delta_IJ alpha(I,J) = alpha parameter for pair potential between I and J (can be computed from bulk modulus of reference structure) re(I,J) = equilibrium distance between I and J in the reference structure Cmax(I,J,K) = Cmax screening parameter when I-J pair is screened by K (I<=J); default = 2.8 Cmin(I,J,K) = Cmin screening parameter when I-J pair is screened by K (I<=J); default = 2.0 lattce(I,J) = lattice structure of I-J reference structure: fcc = face centered cubic bcc = body centered cubic hcp = hexagonal close-packed dim = dimer dia = diamond (interlaced fcc for alloy) dia3= diamond structure with primary 1NN and secondary 3NN interaction b1 = rock salt (NaCl structure) c11 = MoSi2 structure l12 = Cu3Au structure (lower case L, followed by 12) b2 = CsCl structure (interpenetrating simple cubic) ch4 = methane-like structure, only for binary system lin = linear structure (180 degree angle) zig = zigzag structure with a uniform angle tri = H2O-like structure that has an angle sc = simple cubic nn2(I,J) = turn on second-nearest neighbor MEAM formulation for I-J pair (see for example `(Lee) `__). 0 = second-nearest neighbor formulation off 1 = second-nearest neighbor formulation on default = 0 attrac(I,J) = additional cubic attraction term in Rose energy I-J pair potential default = 0 repuls(I,J) = additional cubic repulsive term in Rose energy I-J pair potential default = 0 zbl(I,J) = blend the MEAM I-J pair potential with the ZBL potential for small atom separations `(ZBL) `__ default = 1 theta(I,J) = angle between three atoms in line, zigzag, and trimer reference structures in degrees default = 180 gsmooth_factor = factor determining the length of the G-function smoothing region; only significant for ibar=0 or ibar=4. 99.0 = short smoothing region, sharp step 0.5 = long smoothing region, smooth step default = 99.0 augt1 = integer flag for whether to augment t1 parameter by 3/5\*t3 to account for old vs. new meam formulations; 0 = don't augment t1 1 = augment t1 default = 1 ialloy = integer flag to use alternative averaging rule for t parameters, for comparison with the DYNAMO MEAM code 0 = standard averaging (matches ialloy=0 in DYNAMO) 1 = alternative averaging (matches ialloy=1 in DYNAMO) 2 = no averaging of t (use single-element values) default = 0 mixture_ref_t = integer flag to use mixture average of t to compute the background reference density for alloys, instead of the single-element values (see description and warning elsewhere in this doc page) 0 = do not use mixture averaging for t in the reference density 1 = use mixture averaging for t in the reference density default = 0 erose_form = integer value to select the form of the Rose energy function (see description below). default = 0 emb_lin_neg = integer value to select embedding function for negative densities 0 = F(rho)=0 1 = F(rho) = -asub\*esub\*rho (linear in rho, matches DYNAMO) default = 0 bkgd_dyn = integer value to select background density formula 0 = rho_bkgd = rho_ref_meam(a) (as in the reference structure) 1 = rho_bkgd = rho0_meam(a)\*Z_meam(a) (matches DYNAMO) default = 0 *Rc*, *delr*, *re* are in distance units (Angstroms in the case of metal units). *Ec* and *delta* are in energy units (eV in the case of metal units). Each keyword represents a quantity which is either a scalar, vector, 2d array, or 3d array and must be specified with the correct corresponding array syntax. The indices I,J,K each run from 1 to N where N is the number of MEAM elements being used. Thus these lines rho0(2) = 2.25 alpha(1,2) = 4.37 set *rho0* for the second element to the value 2.25 and set *alpha* for the alloy interaction between elements 1 and 2 to 4.37. The *augt1* parameter is related to modifications in the MEAM formulation of the partial electron density function. In recent literature, an extra term is included in the expression for the third-order density in order to make the densities orthogonal (see for example [(Wang)](Wang2), equation 3d); this term is included in the MEAM implementation in LAMMPS. However, in earlier published work this term was not included when deriving parameters, including most of those provided in the `library.meam` file included with LAMMPS, and to account for this difference the parameter *t1* must be augmented by 3/5\*\*t3\*. If *augt1* = 1, the default, this augmentation is done automatically. When parameter values are fit using the modified density function, as in more recent literature, augt1 should be set to 0. The *mixture_ref_t* parameter is available to match results with those of previous versions of LAMMPS (before January 2011). Newer versions of LAMMPS, by default, use the single-element values of the *t* parameters to compute the background reference density. This is the proper way to compute these parameters. Earlier versions of LAMMPS used an alloy mixture averaged value of *t* to compute the background reference density. Setting *mixture_ref_t* = 1 gives the old behavior. WARNING: using *mixture_ref_t* = 1 will give results that are demonstrably incorrect for second-neighbor MEAM, and non-standard for first-neighbor MEAM; this option is included only for matching with previous versions of LAMMPS and should be avoided if possible. The parameters *attrac* and *repuls*, along with the integer selection parameter *erose_form*, can be used to modify the Rose energy function used to compute the pair potential. This function gives the energy of the reference state as a function of interatomic spacing. The form of this function is: astar = alpha \* (r/re - 1.d0) if erose_form = 0: erose = -Ec\*(1+astar+a3\*(astar\*\*3)/(r/re))\*exp(-astar) if erose_form = 1: erose = -Ec\*(1+astar+(-attrac+repuls/r)\*(astar\*\*3))\*exp(-astar) if erose_form = 2: erose = -Ec\*(1 +astar + a3\*(astar\*\*3))\*exp(-astar) a3 = repuls, astar < 0 a3 = attrac, astar >= 0 Most published MEAM parameter sets use the default values *attrac* = *repulse* = 0. Setting *repuls* = *attrac* = *delta* corresponds to the form used in several recent published MEAM parameter sets, such as [(Valone)](Valone) Then using *meam/ms* pair style the multi-state MEAM (MS-MEAM) method is activated. This requires 6 extra parameters in the MEAM library file, resulting in 25 parameters ordered that are ordered like this: elt, lat, z, ielement, atwt, alpha, b0, b1, b2, b3, b1m, b2m, b3m, alat, esub, asub, t0, t1, t2, t3, t1m, t2m, t3m, rozero, ibar The 6 extra MS-MEAM parameters are *b1m, b2m, b3m, t1m, t2m, t3m*. In the LAMMPS `potentials` folder, compatible files have an \".msmeam\" extension. ------------------------------------------------------------------------ Styles with a *gpu*, *intel*, *kk*, *omp*, or *opt* suffix are functionally the same as the corresponding style without the suffix. They have been optimized to run faster, depending on your available hardware, as discussed on the [Accelerator packages](Speed_packages) page. The accelerated styles take the same arguments and should produce the same results, except for round-off and precision issues. These accelerated styles are part of the GPU, INTEL, KOKKOS, OPENMP, and OPT packages, respectively. They are only enabled if LAMMPS was built with those packages. See the [Build package](Build_package) page for more info. You can specify the accelerated styles explicitly in your input script by including their suffix, or you can use the [-suffix command-line switch](Run_options) when you invoke LAMMPS, or you can use the [suffix](suffix) command in your input script. See the [Accelerator packages](Speed_packages) page for more instructions on how to use the accelerated styles effectively. ------------------------------------------------------------------------ :::: note ::: title Note ::: The default form of the *erose* expression in LAMMPS was corrected in March 2009. The current version is correct, but may show different behavior compared with earlier versions of LAMMPS with the attrac and/or repuls parameters are non-zero. To obtain the previous default form, use *erose_form* = 1 (this form does not seem to appear in the literature). An alternative form (see e.g. [(Lee2)](Lee2)) is available using *erose_form* = 2. :::: ------------------------------------------------------------------------ ## Mixing, shift, table, tail correction, restart, rRESPA info 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 with user-specifiable parameters as described above. This pair style does not support the [pair_modify](pair_modify) *shift*, *table*, and *tail* options. This pair style does not write its information to [binary restart files](restart), since it is stored in potential files. Thus, you need to re-specify the pair_style and pair_coeff commands in an input script that reads a restart file. This pair style can only be used via the *pair* keyword of the [run_style respa](run_style) command. It does not support the *inner*, *middle*, *outer* keywords. ------------------------------------------------------------------------ ## Restrictions The *meam* and *meam/ms* pair styles are provided in the MEAM package. They are only enabled if LAMMPS was built with that package. See the [Build package](Build_package) page for more info. The maximum number of elements, that can be read from the MEAM library file, is determined at compile time. The default is 5. If you need support for more elements, you have to change the the constant \'maxelt\' at the beginning of the file `src/MEAM/meam.h` and update/recompile LAMMPS. There is no limit on the number of atoms types. ## Related commands [pair_coeff](pair_coeff), [pair_style eam](pair_eam), [pair_style meam/spline](pair_meam_spline) ## Default none ------------------------------------------------------------------------ ::: {#Baskes} **(Baskes)** Baskes, Phys Rev B, 46, 2727-2742 (1992). ::: ::: {#Baskes2} **(Baskes2)** Baskes, Phys Rev B, 75, 094113 (2007). ::: ::: {#Gullet} **(Gullet)** Gullet, Wagner, Slepoy, SANDIA Report 2003-8782 (2003). This report may be accessed on-line via [this link](https://download.lammps.org/pdfs/MEAM_report_2003.pdf)\_. ::: ::: {#Lee} **(Lee)** Lee, Baskes, Phys. Rev. B, 62, 8564-8567 (2000). ::: ::: {#Lee2} **(Lee2)** Lee, Baskes, Kim, Cho. Phys. Rev. B, 64, 184102 (2001). ::: ::: {#Valone} **(Valone)** Valone, Baskes, Martin, Phys. Rev. B, 73, 214209 (2006). ::: ::: {#Wang2} **(Wang)** Wang, Van Hove, Ross, Baskes, J. Chem. Phys., 121, 5410 (2004). ::: ::: {#ZBL} **(ZBL)** J.F. Ziegler, J.P. Biersack, U. Littmark, \"Stopping and Ranges of Ions in Matter\", Vol 1, 1985, Pergamon Press. :::