xmd-8.html

一个很好的分子动力学程序

<DT><B>DUP ndup xdisp ydisp zdisp</B><DD><P>The DUP command is used to create particles.  Its need has been
superceded by the newer FILL command.
<P>DUP duplicates the selected particles (see SELECT below) ndup times and
displaces each duplicate in the x, y and z directions by xdisp, ydisp
and zdisp; relative to the previous duplicate.
For example the command
<BLOCKQUOTE><CODE>
<PRE>
DUP 3 1 0 0
</PRE>
</CODE></BLOCKQUOTE>

will duplicate the selected particles three times.  The
first, second and third set are offset from the original particles by (1
0 0), (2 0 0), and (3 0 0) respectively.  At the end of the command all
previously selected particles and the newly created duplicates are
selected.  In the example at the end of this document the DUP command is
used to create a BCC lattice.
<P>
<P>
<DT><B>ECHO { ON | OFF }</B><DD><P>By default every command read by XMD is echoed to the output.  This
echoing can be controlled with the ECHO command.  ECHO OFF will stop
command echoing.  ECHO ON is the default.
<P>
<P>
<DT><B>ERASE file</B><DD><P>Erases the file name <CODE>file</CODE>.  This is useful if you want
to add simulations steps to the end of a file but you want to
first clear the file as shown in this example,
<BLOCKQUOTE><CODE>
<PRE>
erase aucrack.cor
repeat 10
   cmd 1000
   write cor +aucrack.cor
end
</PRE>
</CODE></BLOCKQUOTE>
<P>
<P>
<DT><B>ESAVE nskip file</B><DD><P>Causes MC and CMD simulations to save the energy every nskip steps in
file.  For MC simulation two numbers are saved at each relevant step;
the step number and the total potential energy (in that order).  For the
CMD case four numbers are saved: the step number, the total energy (the
potential plus the kinetic energy), the potential energy , and the
kinetic energy.  Note that the potential energy includes the energies
due to external forces, as specified with the EXTFORCE and EXTSPRING
commands.
<P>
<P>
<DT><B>EUNIT [ ERG | EV | JOULE | K | uname uvalue ]</B><DD><P>Sets the units for energy, either ergs, electron volts, joules, Kelvin
or a user specified name and unit.  uvalue is the ratio of the unit
uname to ergs.  All energies printed by the program (either via the
ESAVE command or the WRITE ENERGY command) will be in the units
specified.  The forces read or written by the EXTFORCE command or the
WRITE FORCE command will be in the specified units per centimeter.  The
spring vectors read by the EXTSPRING command will be the in specified
units per centimeter squared.  If no value is specified on the EUNIT
command line then the current value of uname and uvalue will be printed.
The program default vaule for EUNIT is Kelvin (or put another way, 
the default value is the inverse of Boltzman's constant).
<P>
<P>
<P>
<DT><B>EXTFORCE { CLEAR | [ formula [ formula .. ] ] fx fy fz }</B><DD><P>Places an external force on selected atoms equal to fx, fy, fz.  If the
CLEAR option is specified then forces on all atoms are reset to zero
(&quot; all the atoms &quot; means not only the selected atoms).  The unit
for the forces is the energy unit specified by EUNIT (the default is
ergs) divided by centimeters.
<P>The EXTFORCE contributes to the potential (and total) energy written by
the ESAVE command, but not to the energy written by the WRITE ENERGY
command.  The energy for an atom under an external force is
<P>
<BLOCKQUOTE><CODE>
<PRE>
E = - F (r - r0)
</PRE>
</CODE></BLOCKQUOTE>
<P>where F is the applied force, r is the atom position, and r0 is the atom
position at the time the EXTFORCE command was issued.  Thus, for
instance, this energy is zero if an atom is stationary.  The external
forces created with this command will be stored in the state file (see
the WRITE STATE command).
<P>
<P>The expressions for fx fy and fz can contain the variables x, y and z,
which will equal the coordinates of each atom moved.  Thus you can apply
an external force that is a function of an individual atom's position.
The optional formula expression can initialize one or more variables for
subsequent use in the expressions fx, fy and fz.  See the MOVE command
for an example.
<P>
<P>
<DT><B>EXTSPRING { CLEAR | [ formula [ formula .. ] ] kx ky kz }</B><DD><P>Places an external spring on selected atoms.  If the CLEAR option is
specified then all of the external springs are removed ("all the
external springs" means not only those on selected atoms).  The
direction of the spring's force is parallel to the spring vector (kx,
ky, kz).  Also, only the atom's displacement parallel to the vector
determines the magnitude of the force.  The actual equation used is
<P>
<BLOCKQUOTE><CODE>
<PRE>
</PRE>
</CODE></BLOCKQUOTE>
<P>where F is the force due to the spring, r is the atom position, r0 is
the atom position at the time the EXTSPRING command was issued, k is the
spring vector and is a unit vector parallel to (kx, ky, kz).  The energy
due to external springs is included in the potential (and total) energy
written by the ESAVE command, but not by the WRITE ENERGY command.  The
units of the spring vector are the energy units specified by EUNIT (the
default is ergs) divided by centimeters squared.  The external springs
created with this command will be stored in the state file (see the
WRITE STATE command).  The expressions for kx, ky and kz can contain the
variables x, y and z, which will equal the coordinates of each atom
moved.  Thus you can apply an external spring that is a function of an
individual atom's position.  The optional formula expression can
initialize one or more variables for subsequent use in the expressions
kx, ky and kz.  See the MOVE command for an example.
<P>
<P>
<DT><B>FILL ...</B><DD><P>The FILL family of commands is used for creating lattices.
It can create a regular lattice in any orientation and with
any basis vectors and basis atoms.  Please see the 
<A HREF="xmd-9.html#fill">section on the FILL command</A>.
<P>
<DT><B>FIX { ON | OFF }</B><DD><P>For CMD simulations FIX causes the selected particles to be either fixed
(ON option) or free (OFF option).  If FIX is not specified then the
particles are free.  See the section on 
<A HREF="xmd-6.html#implementation-fix">fixed atoms</A> above.
<P>
<P>
<DT><B>ITEMP SEL temp [X|Y|Z] [X|Y|Z] ..</B><DD><P>Assigns random velocities appropriate to temp to all the particles.  The
assigned velocities are chosen at random but conform to a Boltzmann
distribution.  Normally random values are assigned to all components of
the velocity, x, y and z. However when performing a two or one
dimensional simulation, you will want to assign velocities to specific
components only.  By specifying specific directions such as X, Y or Z
you can assign values to those components only.  The remaining
components will be set to zero.  As an example, if you are running a two
dimensional simulation in the xy plane, the command
<P>
<BLOCKQUOTE><CODE>
<PRE>
ITEMP 100 X Y
</PRE>
</CODE></BLOCKQUOTE>
<P>will assign velocities in the x and y directions, and set the z
component to zero.  If in addition the particles all have the same z
coordinates, then the motion will be confined to the xy plane.  See also
the VELOCITY command.
<P>
<P>
<DT><B>LABEL nlabel</B><DD><P>This command is followed by <CODE>nlabel</CODE> lines of text,
<CODE>nlabel</CODE> must not exceed 8.
The text is used to add notes to simulations files, and is
stored in COR, RCV and state files, as well as written out on plots.
<P>
<P>
<A NAME="command-macro"></A> <DT><B>MACRO name string</B><DD><P>This command assigns a string to the name.  For example,
<BLOCKQUOTE><CODE>
<PRE>
macro fname femelt
esave 10 $fname.e
bsave 10 $fname.b
ssave 10 $fname.s
</PRE>
</CODE></BLOCKQUOTE>

tells XMD to replaces &quot;$fname&quot; with &quot;femelt&quot;,
so that the resulting file names will be femelt.e, femelt.b and
femelt.b.
<P>Macros are replaced upon input before any other processing is done.  One
consequence is that macro substitution is done before the commands are
printed to output, so you will see the substituted macros in the output.
<P>You can have blanks in the macro value, the macro value is the string
comprised of the first non-blank character after the macro name until
the end of the line.  So you can have the following macro
<BLOCKQUOTE><CODE>
<PRE>
macro sbox select box
$sbox  0 0 0   5 10 10
$sbox  5 0 0  10 10 10
</PRE>
</CODE></BLOCKQUOTE>

This will expand to
<BLOCKQUOTE><CODE>
<PRE>
macro sbox select box
select box 0 0 0   5 10 10
select box 5 0 0  10 10 10
</PRE>
</CODE></BLOCKQUOTE>

The macro name can be any sequence of numbers and characters, and can
even start with a number, such as
<BLOCKQUOTE><CODE>
<PRE>
macro 1st out1
macri 2nd out2
</PRE>
</CODE></BLOCKQUOTE>

Sometimes you need to tell XMD where the macro name ends, for example
<BLOCKQUOTE><CODE>
<PRE>
macro fname cu_fit
select type 1
#
#  We want to write COR file to  &quot;cu_fit1.cor&quot;
#
# This will write to &quot;.cor&quot;
#
write cor $fname1.cor
#
# This will write to &quot;cu_fit1.cor&quot;
#
write cor $(fname)1.cor
#
# This will write to &quot;cu_fit.cor&quot;
#
select all
write cor $fname.cor
</PRE>
</CODE></BLOCKQUOTE>

In the first <CODE>write cor</CODE> command, XMD thinks you want the macro
named &quot;fname1&quot;, but there is not such macro, so this macro
name is removed and nothing is put in its place.  
The second <CODE>write cor</CODE> statement places &quot;fname&quot; in
parenthesis, which indicates that this is the macro name.
The third <CODE>write cor</CODE> statement does not need the parenthesis
because XMD knows that macro names do not include periods (.),  so the
macro name must end there.
<P>There are 9 special built-in macros, $1 through $dollar;9,
these represent command lines parameters.  See the section on the
<A HREF="xmd-6.html#implementation-macros">implementation of macros</A>.
<P>
<P>
<A NAME="command-macrof"></A> <DT><B>MACROF name format variable</B><DD><P>This command is similar to MACRO, but is assigns a CALC variable to
a macro.  For instance, if you want to write an individual COR file
for specific steps, 
<BLOCKQUOTE><CODE>
<PRE>
repeat 5
  cmd 40
  macrof fname simul%03d nstep
  write cor $fname.cor
end
</PRE>
</CODE></BLOCKQUOTE>

This would write the files simul040.cor, simul080.cor, simul120.cor,
.. simul200.cor.
The <CODE>format</CODE> parameter follows the standard of the C language
printf() commands and can be either <CODE> %d %i %u %x %f %e </CODE>.
Please see information for the C language printf() command.  Under some 
Unix installations, you can learn more about printf with the
command
<BLOCKQUOTE><CODE>
<PRE>
   man printf
</PRE>
</CODE></BLOCKQUOTE>
<P>
<P>
<DT><B>MASS mass</B><DD><P>Assigns mass to all selected particles (units are atomic mass units).
<P>
<P>
<DT><B>MC nstep temp</B><DD><P>Perform a MC simulation for nstep steps and at temperature temp.  The
course of Monte Carlo simulations is affected by the commands CLAMP and
SIZE.  The command ESAVE affects the output from these simulations.
<P>
<P>
<A NAME="command-move"></A> <DT><B>MOVE [ formula [formula .. ] ] xdisp ydisp zdisp</B><DD><P>Moves the selected particles by xdisp, ydisp and zdisp.  The expressions