Page Menu
Home
c4science
Search
Configure Global Search
Log In
Files
F65173703
compute_orientorder_atom.cpp
No One
Temporary
Actions
Download File
Edit File
Delete File
View Transforms
Subscribe
Mute Notifications
Award Token
Subscribers
None
File Metadata
Details
File Info
Storage
Attached
Created
Sat, Jun 1, 12:24
Size
15 KB
Mime Type
text/x-c
Expires
Mon, Jun 3, 12:24 (2 d)
Engine
blob
Format
Raw Data
Handle
18018595
Attached To
rLAMMPS lammps
compute_orientorder_atom.cpp
View Options
/* ----------------------------------------------------------------------
LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
http://lammps.sandia.gov, Sandia National Laboratories
Steve Plimpton, sjplimp@sandia.gov
Copyright (2003) Sandia Corporation. Under the terms of Contract
DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
certain rights in this software. This software is distributed under
the GNU General Public License.
See the README file in the top-level LAMMPS directory.
------------------------------------------------------------------------- */
/* ----------------------------------------------------------------------
Contributing author: Aidan Thompson (SNL)
Axel Kohlmeyer (Temple U)
------------------------------------------------------------------------- */
#include <string.h>
#include <stdlib.h>
#include "compute_orientorder_atom.h"
#include "atom.h"
#include "update.h"
#include "modify.h"
#include "neighbor.h"
#include "neigh_list.h"
#include "neigh_request.h"
#include "force.h"
#include "pair.h"
#include "comm.h"
#include "memory.h"
#include "error.h"
#include "math_const.h"
using
namespace
LAMMPS_NS
;
using
namespace
MathConst
;
using
namespace
std
;
#ifdef DBL_EPSILON
#define MY_EPSILON (10.0*DBL_EPSILON)
#else
#define MY_EPSILON (10.0*2.220446049250313e-16)
#endif
/* ---------------------------------------------------------------------- */
ComputeOrientOrderAtom
::
ComputeOrientOrderAtom
(
LAMMPS
*
lmp
,
int
narg
,
char
**
arg
)
:
Compute
(
lmp
,
narg
,
arg
),
qlist
(
NULL
),
distsq
(
NULL
),
nearest
(
NULL
),
rlist
(
NULL
),
qnarray
(
NULL
),
qnm_r
(
NULL
),
qnm_i
(
NULL
)
{
if
(
narg
<
3
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
// set default values for optional args
nnn
=
12
;
cutsq
=
0.0
;
qlcompflag
=
0
;
// specify which orders to request
nqlist
=
5
;
memory
->
create
(
qlist
,
nqlist
,
"orientorder/atom:qlist"
);
qlist
[
0
]
=
4
;
qlist
[
1
]
=
6
;
qlist
[
2
]
=
8
;
qlist
[
3
]
=
10
;
qlist
[
4
]
=
12
;
qmax
=
12
;
// process optional args
int
iarg
=
3
;
while
(
iarg
<
narg
)
{
if
(
strcmp
(
arg
[
iarg
],
"nnn"
)
==
0
)
{
if
(
iarg
+
2
>
narg
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
if
(
strcmp
(
arg
[
iarg
+
1
],
"NULL"
)
==
0
)
{
nnn
=
0
;
}
else
{
nnn
=
force
->
numeric
(
FLERR
,
arg
[
iarg
+
1
]);
if
(
nnn
<=
0
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
}
iarg
+=
2
;
}
else
if
(
strcmp
(
arg
[
iarg
],
"degrees"
)
==
0
)
{
if
(
iarg
+
2
>
narg
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
nqlist
=
force
->
numeric
(
FLERR
,
arg
[
iarg
+
1
]);
if
(
nqlist
<=
0
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
memory
->
destroy
(
qlist
);
memory
->
create
(
qlist
,
nqlist
,
"orientorder/atom:qlist"
);
iarg
+=
2
;
if
(
iarg
+
nqlist
>
narg
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
qmax
=
0
;
for
(
int
iw
=
0
;
iw
<
nqlist
;
iw
++
)
{
qlist
[
iw
]
=
force
->
numeric
(
FLERR
,
arg
[
iarg
+
iw
]);
if
(
qlist
[
iw
]
<
0
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
if
(
qlist
[
iw
]
>
qmax
)
qmax
=
qlist
[
iw
];
}
iarg
+=
nqlist
;
}
else
if
(
strcmp
(
arg
[
iarg
],
"components"
)
==
0
)
{
qlcompflag
=
1
;
if
(
iarg
+
2
>
narg
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
qlcomp
=
force
->
numeric
(
FLERR
,
arg
[
iarg
+
1
]);
if
(
qlcomp
<=
0
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
iqlcomp
=
-
1
;
for
(
int
iw
=
0
;
iw
<
nqlist
;
iw
++
)
if
(
qlcomp
==
qlist
[
iw
])
{
iqlcomp
=
iw
;
break
;
}
if
(
iqlcomp
<
0
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
iarg
+=
2
;
}
else
if
(
strcmp
(
arg
[
iarg
],
"cutoff"
)
==
0
)
{
if
(
iarg
+
2
>
narg
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
double
cutoff
=
force
->
numeric
(
FLERR
,
arg
[
iarg
+
1
]);
if
(
cutoff
<=
0.0
)
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
cutsq
=
cutoff
*
cutoff
;
iarg
+=
2
;
}
else
error
->
all
(
FLERR
,
"Illegal compute orientorder/atom command"
);
}
if
(
qlcompflag
)
ncol
=
nqlist
+
2
*
(
2
*
qlcomp
+
1
);
else
ncol
=
nqlist
;
peratom_flag
=
1
;
size_peratom_cols
=
ncol
;
nmax
=
0
;
maxneigh
=
0
;
}
/* ---------------------------------------------------------------------- */
ComputeOrientOrderAtom
::~
ComputeOrientOrderAtom
()
{
memory
->
destroy
(
qnarray
);
memory
->
destroy
(
distsq
);
memory
->
destroy
(
rlist
);
memory
->
destroy
(
nearest
);
memory
->
destroy
(
qlist
);
memory
->
destroy
(
qnm_r
);
memory
->
destroy
(
qnm_i
);
}
/* ---------------------------------------------------------------------- */
void
ComputeOrientOrderAtom
::
init
()
{
if
(
force
->
pair
==
NULL
)
error
->
all
(
FLERR
,
"Compute orientorder/atom requires a "
"pair style be defined"
);
if
(
cutsq
==
0.0
)
cutsq
=
force
->
pair
->
cutforce
*
force
->
pair
->
cutforce
;
else
if
(
sqrt
(
cutsq
)
>
force
->
pair
->
cutforce
)
error
->
all
(
FLERR
,
"Compute orientorder/atom cutoff is "
"longer than pairwise cutoff"
);
memory
->
create
(
qnm_r
,
qmax
,
2
*
qmax
+
1
,
"orientorder/atom:qnm_r"
);
memory
->
create
(
qnm_i
,
qmax
,
2
*
qmax
+
1
,
"orientorder/atom:qnm_i"
);
// need an occasional full neighbor list
int
irequest
=
neighbor
->
request
(
this
,
instance_me
);
neighbor
->
requests
[
irequest
]
->
pair
=
0
;
neighbor
->
requests
[
irequest
]
->
compute
=
1
;
neighbor
->
requests
[
irequest
]
->
half
=
0
;
neighbor
->
requests
[
irequest
]
->
full
=
1
;
neighbor
->
requests
[
irequest
]
->
occasional
=
1
;
int
count
=
0
;
for
(
int
i
=
0
;
i
<
modify
->
ncompute
;
i
++
)
if
(
strcmp
(
modify
->
compute
[
i
]
->
style
,
"orientorder/atom"
)
==
0
)
count
++
;
if
(
count
>
1
&&
comm
->
me
==
0
)
error
->
warning
(
FLERR
,
"More than one compute orientorder/atom"
);
}
/* ---------------------------------------------------------------------- */
void
ComputeOrientOrderAtom
::
init_list
(
int
id
,
NeighList
*
ptr
)
{
list
=
ptr
;
}
/* ---------------------------------------------------------------------- */
void
ComputeOrientOrderAtom
::
compute_peratom
()
{
int
i
,
j
,
ii
,
jj
,
inum
,
jnum
;
double
xtmp
,
ytmp
,
ztmp
,
delx
,
dely
,
delz
,
rsq
;
int
*
ilist
,
*
jlist
,
*
numneigh
,
**
firstneigh
;
invoked_peratom
=
update
->
ntimestep
;
// grow order parameter array if necessary
if
(
atom
->
nmax
>
nmax
)
{
memory
->
destroy
(
qnarray
);
nmax
=
atom
->
nmax
;
memory
->
create
(
qnarray
,
nmax
,
ncol
,
"orientorder/atom:qnarray"
);
array_atom
=
qnarray
;
}
// invoke full neighbor list (will copy or build if necessary)
neighbor
->
build_one
(
list
);
inum
=
list
->
inum
;
ilist
=
list
->
ilist
;
numneigh
=
list
->
numneigh
;
firstneigh
=
list
->
firstneigh
;
// compute order parameter for each atom in group
// use full neighbor list to count atoms less than cutoff
double
**
x
=
atom
->
x
;
int
*
mask
=
atom
->
mask
;
for
(
ii
=
0
;
ii
<
inum
;
ii
++
)
{
i
=
ilist
[
ii
];
double
*
qn
=
qnarray
[
i
];
if
(
mask
[
i
]
&
groupbit
)
{
xtmp
=
x
[
i
][
0
];
ytmp
=
x
[
i
][
1
];
ztmp
=
x
[
i
][
2
];
jlist
=
firstneigh
[
i
];
jnum
=
numneigh
[
i
];
// insure distsq and nearest arrays are long enough
if
(
jnum
>
maxneigh
)
{
memory
->
destroy
(
distsq
);
memory
->
destroy
(
rlist
);
memory
->
destroy
(
nearest
);
maxneigh
=
jnum
;
memory
->
create
(
distsq
,
maxneigh
,
"orientorder/atom:distsq"
);
memory
->
create
(
rlist
,
maxneigh
,
3
,
"orientorder/atom:rlist"
);
memory
->
create
(
nearest
,
maxneigh
,
"orientorder/atom:nearest"
);
}
// loop over list of all neighbors within force cutoff
// distsq[] = distance sq to each
// rlist[] = distance vector to each
// nearest[] = atom indices of neighbors
int
ncount
=
0
;
for
(
jj
=
0
;
jj
<
jnum
;
jj
++
)
{
j
=
jlist
[
jj
];
j
&=
NEIGHMASK
;
delx
=
xtmp
-
x
[
j
][
0
];
dely
=
ytmp
-
x
[
j
][
1
];
delz
=
ztmp
-
x
[
j
][
2
];
rsq
=
delx
*
delx
+
dely
*
dely
+
delz
*
delz
;
if
(
rsq
<
cutsq
)
{
distsq
[
ncount
]
=
rsq
;
rlist
[
ncount
][
0
]
=
delx
;
rlist
[
ncount
][
1
]
=
dely
;
rlist
[
ncount
][
2
]
=
delz
;
nearest
[
ncount
++
]
=
j
;
}
}
// if not nnn neighbors, order parameter = 0;
if
((
ncount
==
0
)
||
(
ncount
<
nnn
))
{
for
(
int
iw
=
0
;
iw
<
nqlist
;
iw
++
)
qn
[
iw
]
=
0.0
;
continue
;
}
// if nnn > 0, use only nearest nnn neighbors
if
(
nnn
>
0
)
{
select3
(
nnn
,
ncount
,
distsq
,
nearest
,
rlist
);
ncount
=
nnn
;
}
calc_boop
(
rlist
,
ncount
,
qn
,
qlist
,
nqlist
);
}
}
}
/* ----------------------------------------------------------------------
memory usage of local atom-based array
------------------------------------------------------------------------- */
double
ComputeOrientOrderAtom
::
memory_usage
()
{
double
bytes
=
ncol
*
nmax
*
sizeof
(
double
);
bytes
+=
(
qmax
*
(
2
*
qmax
+
1
)
+
maxneigh
*
4
)
*
sizeof
(
double
);
bytes
+=
(
nqlist
+
maxneigh
)
*
sizeof
(
int
);
return
bytes
;
}
/* ----------------------------------------------------------------------
select3 routine from Numerical Recipes (slightly modified)
find k smallest values in array of length n
sort auxiliary arrays at same time
------------------------------------------------------------------------- */
// Use no-op do while to create single statement
#define SWAP(a,b) do { \
tmp = a; a = b; b = tmp; \
} while(0)
#define ISWAP(a,b) do { \
itmp = a; a = b; b = itmp; \
} while(0)
#define SWAP3(a,b) do { \
tmp = a[0]; a[0] = b[0]; b[0] = tmp; \
tmp = a[1]; a[1] = b[1]; b[1] = tmp; \
tmp = a[2]; a[2] = b[2]; b[2] = tmp; \
} while(0)
/* ---------------------------------------------------------------------- */
void
ComputeOrientOrderAtom
::
select3
(
int
k
,
int
n
,
double
*
arr
,
int
*
iarr
,
double
**
arr3
)
{
int
i
,
ir
,
j
,
l
,
mid
,
ia
,
itmp
;
double
a
,
tmp
,
a3
[
3
];
arr
--
;
iarr
--
;
arr3
--
;
l
=
1
;
ir
=
n
;
for
(;;)
{
if
(
ir
<=
l
+
1
)
{
if
(
ir
==
l
+
1
&&
arr
[
ir
]
<
arr
[
l
])
{
SWAP
(
arr
[
l
],
arr
[
ir
]);
ISWAP
(
iarr
[
l
],
iarr
[
ir
]);
SWAP3
(
arr3
[
l
],
arr3
[
ir
]);
}
return
;
}
else
{
mid
=
(
l
+
ir
)
>>
1
;
SWAP
(
arr
[
mid
],
arr
[
l
+
1
]);
ISWAP
(
iarr
[
mid
],
iarr
[
l
+
1
]);
SWAP3
(
arr3
[
mid
],
arr3
[
l
+
1
]);
if
(
arr
[
l
]
>
arr
[
ir
])
{
SWAP
(
arr
[
l
],
arr
[
ir
]);
ISWAP
(
iarr
[
l
],
iarr
[
ir
]);
SWAP3
(
arr3
[
l
],
arr3
[
ir
]);
}
if
(
arr
[
l
+
1
]
>
arr
[
ir
])
{
SWAP
(
arr
[
l
+
1
],
arr
[
ir
]);
ISWAP
(
iarr
[
l
+
1
],
iarr
[
ir
]);
SWAP3
(
arr3
[
l
+
1
],
arr3
[
ir
]);
}
if
(
arr
[
l
]
>
arr
[
l
+
1
])
{
SWAP
(
arr
[
l
],
arr
[
l
+
1
]);
ISWAP
(
iarr
[
l
],
iarr
[
l
+
1
]);
SWAP3
(
arr3
[
l
],
arr3
[
l
+
1
]);
}
i
=
l
+
1
;
j
=
ir
;
a
=
arr
[
l
+
1
];
ia
=
iarr
[
l
+
1
];
a3
[
0
]
=
arr3
[
l
+
1
][
0
];
a3
[
1
]
=
arr3
[
l
+
1
][
1
];
a3
[
2
]
=
arr3
[
l
+
1
][
2
];
for
(;;)
{
do
i
++
;
while
(
arr
[
i
]
<
a
);
do
j
--
;
while
(
arr
[
j
]
>
a
);
if
(
j
<
i
)
break
;
SWAP
(
arr
[
i
],
arr
[
j
]);
ISWAP
(
iarr
[
i
],
iarr
[
j
]);
SWAP3
(
arr3
[
i
],
arr3
[
j
]);
}
arr
[
l
+
1
]
=
arr
[
j
];
arr
[
j
]
=
a
;
iarr
[
l
+
1
]
=
iarr
[
j
];
iarr
[
j
]
=
ia
;
arr3
[
l
+
1
][
0
]
=
arr3
[
j
][
0
];
arr3
[
l
+
1
][
1
]
=
arr3
[
j
][
1
];
arr3
[
l
+
1
][
2
]
=
arr3
[
j
][
2
];
arr3
[
j
][
0
]
=
a3
[
0
];
arr3
[
j
][
1
]
=
a3
[
1
];
arr3
[
j
][
2
]
=
a3
[
2
];
if
(
j
>=
k
)
ir
=
j
-
1
;
if
(
j
<=
k
)
l
=
i
;
}
}
}
/* ----------------------------------------------------------------------
calculate the bond orientational order parameters
------------------------------------------------------------------------- */
void
ComputeOrientOrderAtom
::
calc_boop
(
double
**
rlist
,
int
ncount
,
double
qn
[],
int
qlist
[],
int
nqlist
)
{
for
(
int
iw
=
0
;
iw
<
nqlist
;
iw
++
)
{
int
n
=
qlist
[
iw
];
qn
[
iw
]
=
0.0
;
for
(
int
m
=
0
;
m
<
2
*
n
+
1
;
m
++
)
{
qnm_r
[
iw
][
m
]
=
0.0
;
qnm_i
[
iw
][
m
]
=
0.0
;
}
}
for
(
int
ineigh
=
0
;
ineigh
<
ncount
;
ineigh
++
)
{
const
double
*
const
r
=
rlist
[
ineigh
];
double
rmag
=
dist
(
r
);
if
(
rmag
<=
MY_EPSILON
)
{
return
;
}
double
costheta
=
r
[
2
]
/
rmag
;
double
expphi_r
=
r
[
0
];
double
expphi_i
=
r
[
1
];
double
rxymag
=
sqrt
(
expphi_r
*
expphi_r
+
expphi_i
*
expphi_i
);
if
(
rxymag
<=
MY_EPSILON
)
{
expphi_r
=
1.0
;
expphi_i
=
0.0
;
}
else
{
double
rxymaginv
=
1.0
/
rxymag
;
expphi_r
*=
rxymaginv
;
expphi_i
*=
rxymaginv
;
}
for
(
int
iw
=
0
;
iw
<
nqlist
;
iw
++
)
{
int
n
=
qlist
[
iw
];
qnm_r
[
iw
][
n
]
+=
polar_prefactor
(
n
,
0
,
costheta
);
double
expphim_r
=
expphi_r
;
double
expphim_i
=
expphi_i
;
for
(
int
m
=
1
;
m
<=
+
n
;
m
++
)
{
double
prefactor
=
polar_prefactor
(
n
,
m
,
costheta
);
double
c_r
=
prefactor
*
expphim_r
;
double
c_i
=
prefactor
*
expphim_i
;
qnm_r
[
iw
][
m
+
n
]
+=
c_r
;
qnm_i
[
iw
][
m
+
n
]
+=
c_i
;
if
(
m
&
1
)
{
qnm_r
[
iw
][
-
m
+
n
]
-=
c_r
;
qnm_i
[
iw
][
-
m
+
n
]
+=
c_i
;
}
else
{
qnm_r
[
iw
][
-
m
+
n
]
+=
c_r
;
qnm_i
[
iw
][
-
m
+
n
]
-=
c_i
;
}
double
tmp_r
=
expphim_r
*
expphi_r
-
expphim_i
*
expphi_i
;
double
tmp_i
=
expphim_r
*
expphi_i
+
expphim_i
*
expphi_r
;
expphim_r
=
tmp_r
;
expphim_i
=
tmp_i
;
}
}
}
double
fac
=
sqrt
(
MY_4PI
)
/
ncount
;
double
normfac
=
0.0
;
for
(
int
iw
=
0
;
iw
<
nqlist
;
iw
++
)
{
int
n
=
qlist
[
iw
];
double
qm_sum
=
0.0
;
for
(
int
m
=
0
;
m
<
2
*
n
+
1
;
m
++
)
{
qm_sum
+=
qnm_r
[
iw
][
m
]
*
qnm_r
[
iw
][
m
]
+
qnm_i
[
iw
][
m
]
*
qnm_i
[
iw
][
m
];
// printf("Ylm^2 = %d %d %g\n",n,m,
// qnm_r[iw][m]*qnm_r[iw][m] + qnm_i[iw][m]*qnm_i[iw][m]);
}
qn
[
iw
]
=
fac
*
sqrt
(
qm_sum
/
(
2
*
n
+
1
));
if
(
qlcompflag
&&
iqlcomp
==
iw
)
normfac
=
1.0
/
sqrt
(
qm_sum
);
}
// output of the complex vector
if
(
qlcompflag
)
{
int
j
=
nqlist
;
for
(
int
m
=
0
;
m
<
2
*
qlcomp
+
1
;
m
++
)
{
qn
[
j
++
]
=
qnm_r
[
iqlcomp
][
m
]
*
normfac
;
qn
[
j
++
]
=
qnm_i
[
iqlcomp
][
m
]
*
normfac
;
}
}
}
/* ----------------------------------------------------------------------
calculate scalar distance
------------------------------------------------------------------------- */
double
ComputeOrientOrderAtom
::
dist
(
const
double
r
[])
{
return
sqrt
(
r
[
0
]
*
r
[
0
]
+
r
[
1
]
*
r
[
1
]
+
r
[
2
]
*
r
[
2
]);
}
/* ----------------------------------------------------------------------
polar prefactor for spherical harmonic Y_l^m, where
Y_l^m (theta, phi) = prefactor(l, m, cos(theta)) * exp(i*m*phi)
------------------------------------------------------------------------- */
double
ComputeOrientOrderAtom
::
polar_prefactor
(
int
l
,
int
m
,
double
costheta
)
{
const
int
mabs
=
abs
(
m
);
double
prefactor
=
1.0
;
for
(
int
i
=
l
-
mabs
+
1
;
i
<
l
+
mabs
+
1
;
++
i
)
prefactor
*=
static_cast
<
double
>
(
i
);
prefactor
=
sqrt
(
static_cast
<
double
>
(
2
*
l
+
1
)
/
(
MY_4PI
*
prefactor
))
*
associated_legendre
(
l
,
mabs
,
costheta
);
if
((
m
<
0
)
&&
(
m
%
2
))
prefactor
=
-
prefactor
;
return
prefactor
;
}
/* ----------------------------------------------------------------------
associated legendre polynomial
------------------------------------------------------------------------- */
double
ComputeOrientOrderAtom
::
associated_legendre
(
int
l
,
int
m
,
double
x
)
{
if
(
l
<
m
)
return
0.0
;
double
p
(
1.0
),
pm1
(
0.0
),
pm2
(
0.0
);
if
(
m
!=
0
)
{
const
double
sqx
=
sqrt
(
1.0
-
x
*
x
);
for
(
int
i
=
1
;
i
<
m
+
1
;
++
i
)
p
*=
static_cast
<
double
>
(
2
*
i
-
1
)
*
sqx
;
}
for
(
int
i
=
m
+
1
;
i
<
l
+
1
;
++
i
)
{
pm2
=
pm1
;
pm1
=
p
;
p
=
(
static_cast
<
double
>
(
2
*
i
-
1
)
*
x
*
pm1
-
static_cast
<
double
>
(
i
+
m
-
1
)
*
pm2
)
/
static_cast
<
double
>
(
i
-
m
);
}
return
p
;
}
Event Timeline
Log In to Comment