📄 stress.xmo
字号:
XMD (Version 2.5.29 Nov 27 2000)Not using pthread library.Not using asm("finit"); patch in thread routines.######################################################################## ## Example: Calculate the internal stress ## Use this to find lattice parameter at a finite ## temperature. ## ## NOTE: This method is obsolete. Instead, see xmd documentaiton ## about the PRESSURE CLAMP command. (JR 30 Jun 1999). ## ######################################################################### ## In order to do simulations at a finite temperature, one needs ## to know what the lattice parameters are at that temperature. (At ## this point, we are including all lattice, such as tetragonal and ## monoclinic, which can have up to 6 lattice parameters. Later on ## we will restrict the discussion to cubic lattice, which have only ## one lattice parameter). This can be done by plotting the time ## averaged internal stress of a lattice versus the lattice ## parameters. From this plot one can interpolate the value of the ## lattice parameter that gives zero stress. ## ## Here we show how to calculate the internal stress for at ## lattice at a finite temperature and for one value of the lattice ## parameter. This example uses a cubic NiAl lattice, so there is ## only one lattice parameter, which we choose to be 2.9 angstroms. ## We run this simulation with a CLAMP value of 300K. The stress ## values are calculated and written to a file using the SSAVE ## command, which is analogous to the ESAVE command. The command ## ## SSAVE 1 stress.str ## ## saves the stresses in a text file called <stress.str>. The ## first column of this file holds the step number, the next six ## columns holds the Voight stresses, s1 through s6. ## ## After the run, each column of stress values should be plotted ## versus time step. During the early time steps the stresses will ## be equilibrating. After this initial period the stresses should ## settle into a period of steady fluctuation. The time averages of ## the stresses after the initial period should be calculated. ## These average stresses can now be plotted on a different graph as ## a function of lattice constant. ## ## With a cubic lattice, as we have in our example, symmetry ## dictates that on average, s1, s2 and s3 will be equal and s4, s5 ## and s6 will be zero. Thus instead of plotting 6 stress in our ## second graph, we need only plot one, the average of s1, s2 and ## s3. The other three can be taken to be zero. ## ## We then repeat the process for another value of the lattice ## parameters, producing a stress versus time step plot from each ## run, which in turn produces new points to add to the second plot, ## the average stress versus lattice parameter. ## ######################################################################### Set cubic lattice parametercalc A0 = 2.9# Read potential for nialread ../nial.txt# Make room for all particlesallocate 500 15000WARNING: Allocate command is now obsolete.# Make repeating box and lattice (in units of a0)box 6 6 6particle 2dup 5 1 0 0dup 5 0 1 0dup 5 0 0 1# Scale up to units of angstroms (2.8712 unit cell)scale A0# Save stresses from every dynamics step in file "stress.str"ssave 1 stress.str# Set particle masses (in atomic mass units)select type 1*** NUMBER SELECTED 216mass 58.71select type 2*** NUMBER SELECTED 216mass 26.982dtime 3.5e-15# Set adiabtic simulation at starting temperature of 200Kclamp 300itemp 300# Perform dynamicscmd 100*** Current step is 100TIME INFORMATION Ending Time: 03/19/2001 13:41:09 Elapsed Time: 0 min 06 secDYNAMICS STATISTICS Number of MD Steps: 100 Number of Neighbor Searchs: 8 Time spent on MD steps: 5 (secs) Time spent on Neighbor Search: 0 (secs)ERROR STATISTICS Number of Fatal Errors: 0 Number of Unknown Command Errors: 0 Number of Misc. Warnings: 1 <======= !!!
⌨️ 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 + -