📄 xmd-8.html
字号:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"><HTML><HEAD> <META NAME="GENERATOR" CONTENT="SGML-Tools 1.0.9"> <TITLE>XMD - Molecular Dynamics Program: Commands</TITLE> <LINK HREF="xmd-9.html" REL=next> <LINK HREF="xmd-7.html" REL=previous> <LINK HREF="xmd.html#toc8" REL=contents></HEAD><BODY><A HREF="xmd-9.html">Next</A><A HREF="xmd-7.html">Previous</A><A HREF="xmd.html#toc8">Contents</A><HR><H2><A NAME="s8">8. Commands</A></H2><P><P><DL><P><DT><B> \ (backslash)</B><DD><P>This is not really a command, but a special character which can be usedto extend a command onto multiple lines. When the last character in acommand is a " \ " , then the next line appended to the end ofthe current line. A command can span multiple lines, but it is limitedto 1024 characters (typically between 12 and 20 lines).<P><DT><B> # (pound sign) </B><DD><P><P><P><DT><B> * (asterisk) </B><DD><P>Used for commenting instruction files. Lines whose first non-blankcharacter is an asterisk or a pound sign will be ignored.<P><P><A NAME="command-box"></A> <DT><B>BOX [SCALE] xbox ybox zbox</B><DD><P>Specifies the box size (in angstroms). The box size is relevant when(1) performing a calculation with repeating boundary conditions (seeoption SURFACE) and (2) when writing a RCV file (see RCV). If theoption scale is chosen then any existing coordinates will be re-scaled(relative to the previous box size) to fit the new box.<P><P><DT><B>BSAVE nskip file</B><DD><P>Causes CMD simulations to save the step number and the x, y and z boxsizes every nskip steps in file. This is useful for monitering the boxsize when using the PRESSURE command which allows the box size tochange.<P><P><A NAME="command-calc"></A> <DT><B>CALC {expression..} </B><DD><P>CALC is a rich command which provides a simple computational language.For instance if the command<P><BLOCKQUOTE><CODE><PRE>CALC x = 2^(1/3)</PRE></CODE></BLOCKQUOTE><P>is given, then a variable x is created that is set equal to 2 to the 1/3power. Later one could have a command<P><BLOCKQUOTE><CODE><PRE>SCALE x</PRE></CODE></BLOCKQUOTE><P>which would scale all the particles (and box) by 2^(1/3). This variable can also be written out with thestatement<P><BLOCKQUOTE><CODE><PRE>WRITE X</PRE></CODE></BLOCKQUOTE><P>(see the WRITE string command below).Furthermore in all commands which require a number, an algebraicexpression may be used (provided there are no embedded blanks). Thusyou could also have the command<P><BLOCKQUOTE><CODE><PRE>SCALE 2^(1/3)</PRE></CODE></BLOCKQUOTE><P>The allowed operators are: + - * / ^ ( ) =<P>There are also allowed functions which are: sin() cos() tan() exp()log() log10() acos() asin() atan() abs() sqrt() int() rand().The functions sin(x), cos(x) and tan(x) expect angle x to be in radians.The functions asin(x), acos(x) and atan(x) return the angle in units ofradians.The function exp(x) return e^x.The function log(x) returns log base e of x,log10(x) returns log base 10.The function abs(x) returns the absolute value of x.The function sqrt(x) returns the square root of x.The function int(x) returns the integer portion of x.The function rand(x) returns a random number from the uniformdistribution between 0 and x.<P>There are two built in constants: pi and e.<P>There can be up to 128 variables set created in one program.See more on CALC under the section on <A HREF="xmd-6.html#implementation-calc">implementation</A>.<P><P><DT><B>CLAMP [SEL] {temp [cstep] | OFF }</B><DD><P>Maintains the temperature for both MC and CMD simulations. For the MC<CODE>temp</CODE> is used as the Boltzmann temperature and cstep is ignored.<P>For the CMD the particles velocities are scaled by(T/<CODE>temp</CODE>)(1/(2*<CODE>cstep</CODE>))at each CMD time step. Here T is the instantaneous system temperature.The application of this factor has the effect of forcing the particlevelocities to a value appropriate for the temperature <CODE>temp</CODE>.The parameter cstep is used to control the rapidity at which the targettemperature is approached. If temp is set to -1, then no temperatureclamp is used (this would be an adiabatic system). See section onTemperature Control.<P>XMD can maintain up to 4 separate particle groups at differenttemperatures. When the SEL option is used, the <CODE>temp</CODE> and<CODE>cstep</CODE> or the OFF setting is applied to the selected particles,the CLAMP settings for all other particles remain unchanged.When the CLAMP command is given without the SEL option, then allparticles are set the same.<P><P><DT><B> CLAMP INFO</B><DD><P>This prints information about the up to 4 separate clamp settings.<P><P><DT><B>CMD nstep</B><DD><P>Initiates a CMD simulation for nstep time steps.If ESAVE, BSAVE, SSAVE or TSAVE have beenimplemented then the corresponding data will be written to disk.Commands which directly affect the course of the CMD simulation areCLAMP, CONSTRAIN, DAMP, DTIME, EXTFORCE, EXTSPRING, FIX, MASS, and SURFACE;as well as the particle types and coordinates as determined by PARTICLE,TYPE and STATE; and the interatomic potential as determined by POTSTATEand POTENTIAL commands.<P><P><DT><B>CONSTRAIN OFF | {LINE | PLANE} xdir ydir zdir xpnt ypnt zpnt</B><DD><P>Applies (LINE or PLANE) or removes (OFF) a constraint to the selectedparticles. The LINE and PLANE constraints force a particle to remain ona line or plane, regardless of the forces it experiences. In practicethis is done by ignoring any forces which are normal to the line orplane. The values xdir, ydir, zdir specify either the line direction orthe plane normal. The values xpnt, ypnt, zpnt specify a coordinatewithin the line or plane (it can be any one of many coordinates) - thisis necessary to locate the line or plane somewhere in space.<P><P><DT><B>CONSTRAIN CAVITY ELLIPSOID spring xcenter ycenter zcenter xaxisyaxis zaxis</B><DD><P><P><P><DT><B>CONSTRAIN CAVITY SPHERE spring xcenter ycenter zcenterradius</B><DD><P>This command places an ellipsoidal " cavity " in thesimulation. The cavity walls reflect particles. When a particle passesinto a wall it experiences a spring force pushing back out. The forceis equal to<BLOCKQUOTE><CODE><PRE>F = 1/2 * spring * dq ^2</PRE></CODE></BLOCKQUOTE>where<CODE>dq</CODE>is the normal distance from wall to the particle, and<CODE>spring</CODE>is the spring constant. Once the particle passes back out of the wall,it no longer feels the force. Only one cavity can be present in asimulation. All particles will be affected by the wall.(xcenter,ycenter,zcenter) specifies the center of the cavity. For aspherical cavity, radius is the radius. For an ellipsoidal cavity,(xaxis,yaxis,zaxis) are the x,y,z half-axis, analogous to sphericalradii. They measure the distance from the ellipsoid center to theellipsoid wall in the x,y,z directions.Please note the following things about the CONSTRAIN CAVITY command.<UL><LI>(1) Only one cavity can be present in a simulation.</LI><LI>(2) The cavity is affected by the SCALE command. If the SCALE commandscales non-uniformly (difference scale in x, y and z directions) thenthe an initially spherical cavity will become an ellipsoid, thesame as if it had been made with the CONSTRAIN CAVITY ELLIPSOID commandabove.</LI><LI>(3) The cavity does know about repeating boundary conditions, so it canoverlap the box.</LI><LI>(4) The ellipsoid cavity can only be orientated with its axis along thex, y and z directions.</LI><LI>(5) A spring constant that is too large will cause instability in theintegration algorithm, the same kind of instability as having a timestep too large. As a rough guide, make sure that spring satisfies thefollowing relation,<BLOCKQUOTE><CODE><PRE>k = 4 pi^2 m / t^2</PRE></CODE></BLOCKQUOTE>wherek is the spring constant,m is the mass in grams of the lightest particle in the simulation,and t is the time step size in seconds.</LI></UL><P><P><DT><B>COR file [ [run] step]</B><DD><P>Reads a COR step from file. If step is specified then the data for thatstep is read; otherwise the first step in the file is read. If step isnot in the file then an error message is given and the programcontinues. If run is specified (which implies that step is alsospecified) then only the data that belongs to the run and step are read.If step is specified without run, and two COR steps have the same stepnumber, then only the first is read. When both run and step arespecified and no matching COR step is found, then an error message isgiven and the program continues.<P>The COR command differs from the RCV command mainly because is reads inparticle types as well as coordinates.<P><P><DT><B>DAMP { OFF | ON [ damp ] }</B><DD><P>Each atom has its own individual velocity damping coefficient (seesection 6). If ON is specified along with damp then specified dampingcoefficient is applied to the selected atoms, and damping will beperformed for each dynamics step. If damp is not specified, the atomicdamping coefficients will not be changed, and damping will be performedwith each dynamics step. If OFF is specified, then damping will not beperformed, but the previous set of damping coefficients will beremembered, and can be applied by a subsequent DAMP ON command.<P><P>The expression for damp can contain the variables x, y and z, which willequal the coordinates of each atom. Thus you can apply a damping termthat is a function of individual atom's position. The optional formulaexpression can initialize one or more variables for subsequent use inthe expressions damp. See the MOVE command for an example.<P><P><A NAME="command-disp"></A> <DT><B> DISP [SEL] { CLEAR | MOVE [scale] | READ n | REFCALC | SCALE scale }</B><DD><P>This command manipulates the displacements. Each particle can have adisplacement associated with it. These displacements can be used tomove the particles. Displacements can be written out using the WRITEDISP command (see the WRITE command). You cannot write out the displacements until you use the DISP commandto create them.<P><P>Note however, that the results of the PLOT DISP command is not affected by this command.For the PLOT DISP command the displacements are calculated independently.<P><DL><P><DT><B>CLEAR</B><DD><P>CLEAR resets displacements to zero. If SEL is specified, only those selected particles have theirdisplacements reset. Without the SEL option, all displacements are reset (and the memory required fordisplacements is released).<P><P><DT><B>MOVE</B><DD><P>MOVE option adds the displacements to the current particles. The SEL option causes the selectedparticles to be moved.<P><P><DT><B>READ</B><DD><P>READ n reads the value of displacements from the input. The displacements should follow on the nextline, with enough values for each coordinate of all the particles (or the selected particles if SEL is set).<P><P><DT><B>REFCALC</B><DD><P>REFCALC set the displacement to the difference between the current particles and the reference particles(current - reference). If SEL is set, only those displacements are affected.<P><P><DT><B>SCALE</B><DD><P>SCALE scale multiplies all the displacements (or the selected displacements if SEL is set) by the amountscale<P></DL><P><P><P><DT><B>DTIME dtime</B><DD><P>This sets the value of the CMD time step in units of seconds. Theoptimum value of the time step is small enough to result in a stableintegration of forces but large enough to provide an efficient use ofcomputer time. See section on techniques for determining the optimumvalue of dtime.<P><P>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -