# Open SourceMolecular Dynamics

### Site Tools

exercises:2017_ethz_mmm:single_point_calculation

# Differences

This shows you the differences between two versions of the page.

 — exercises:2017_ethz_mmm:single_point_calculation [2017/02/22 10:01] (current) Line 1: Line 1: + ====== Computation of the Lennard Jones curve ====== + + In this exercise you will compute the Lennard-Jones energy curve for a system of two Krypton (Kr) atoms.\\ + In Part I you find the instructions for computing the energy of two Kr atoms at a distance $r=4.00 Å$.\\ + In Part II you find the instructions for getting the energy profile as a function of $r$.\\ + Additonal parameters for Neon (Ne) and combination rules to obtain new parameters are provided in  Part III and IV. + + ===== Part I:  Single Point (Energy) calculation ===== + + In this section a commented CP2K input example for a single point calculation is provided. + Comments are added and signaled with '​!'​. + + === 1. Step === + Save the following input to a file named ''​energy.inp''​ + + ​ + &​GLOBAL ​                 ! section to select the kind of calculation + ​RUN_TYPE ENERGY ​      ! select type of calculation. In this case: ENERGY (=Single point calculation) + &END GLOBAL + &​FORCE_EVAL ​             ! section with parameters and system description + METHOD FIST            ! Molecular Mechanics method + &​MM ​                   ! specification of MM parameters ​ + &​FORCEFIELD ​         ! parameters needed to describe the potential ​ + &SPLINE + EMAX_SPLINE 10000    ! numeric parameter to ensure calculation stability. Should not be changed + &END + &​NONBONDED ​      ! parameters for the non bonded interactions + &​LENNARD-JONES ! Lennard-Jones parameters + atoms Kr Kr + EPSILON ​   [K_e] 164.56 + SIGMA [angstrom] ​  3.601 + RCUT  [angstrom] ​ 25.0 + &END LENNARD-JONES + &END NONBONDED + &CHARGE + ATOM Kr + CHARGE 0.0 + &END CHARGE + &END FORCEFIELD + &​POISSON ​             ! solver for non periodic calculations + ​PERIODIC NONE + &EWALD + EWALD_TYPE none + &END EWALD + &END POISSON + &END MM + &​SUBSYS ​                ! system description ​ + &CELL + ABC [angstrom] 10 10 10  ​ + ​PERIODIC NONE + &END CELL + &​COORD ​               ​ + UNIT angstrom + Kr  0 0 0 + Kr  4 0 0 + &END COORD + &​END SUBSYS + &END FORCE_EVAL + ​ + + === 2. Step === + Submit a CP2K calculation with the following commands: + <​code>​ + bsub -n 1  mpirun cp2k.popt -i energy.inp -o energy.out + ​ + + === 3. Step === + Afterwards the file energy.out will look like this: + <​code>​ + **** **** ****** ​ **  PROGRAM STARTED AT                2014-01-20 11:​32:​08.142 + ***** ** ***  *** **   ​PROGRAM STARTED ON                   ​some_server.ethz.ch + ​** ​   ****   ​****** ​   PROGRAM STARTED BY                                   you + ***** **    ** ** **   ​PROGRAM PROCESS ID                                 24183 + **** **  ******* ​ **  PROGRAM STARTED IN                     /​home/​you/​XERCISES + + ... + some stufff + ... + + ENERGY| Total FORCE_EVAL ( FIST ) energy (a.u.): ​          ​0.003617048870059 + ... + some other stuff + ... + **** **** ****** ​ **  PROGRAM ENDED AT                 ​2014-01-20 12:​24:​18.154 + ***** ** ***  *** **   ​PROGRAM RAN ON                       ​some_server.ethz.ch + ​** ​   ****   ​****** ​   PROGRAM RAN BY                                       you + ***** **    ** ** **   ​PROGRAM PROCESS ID                                 24993 + **** **  ******* ​ **  PROGRAM STOPPED IN                   /​home/​you/​EXERCISES + ​ + + If you get the closing Banner you know that cp2k worked. The following line tells you the result: + <​code>​ + ENERGY| Total FORCE_EVAL ( FIST ) energy (a.u.): ​             0.003617048870059 + ​ + + This is the energy (in Hartree) for a system of 2 Kr atoms at distance $r=4.00 Å$ + + Note, that in the input-file ''​EPSILON''​ is given in units of //Kelvin//, whereas in the output the energy is printed in //​Hartree//,​ which is the unit of energy in the system of atomic units (a.u.). To convert from //Kelvin// to //Hartree// you have to multiply with the Boltzmann constant $k_\text{b} = 3.1668154 \cdot 10^{-6} \frac{E_\text{H}}{\text{K}}$ . + + ===== Part II: Computation of the LJ energy curve ===== + + In this section a few scripts to get the LJ energy profiles are presented. + + === 1. Step === + In order to get a good profile, a set of energy values as a function of the interatomic distance is needed. You can use the ''​energy.inp''​ input file and change the Kr coordinates in order to get different starting distances. + + ​ + The output file will be rewritten every time you run a calculation,​ unless you change its name. + ​ + + To do so: + <​code>​ + $mv energy.out energy_dist4A.out + ​ + + + If you run multiple calculations,​ it is always good to keep track of what you have done by producing an input and an output for every distance you are planning to run. + ​ + + For doing so: + <​code>​ +$ cp energy.inp energy_dist2A.inp ​ + ​ + then edit the input file with the new coordinates (e.g. 2 Å). + you can now run CP2K and produce the output file: + <​code>​ + $cp2k.popt -i energy_dist2A.inp -o energy_dist2A.out + ​ + + === 2. Step === + When you have tested a few distances, you can produce a table looking like: + + ^ Input file ^ Distance (Å) ^ Energy ​ ​(Eh) ​ ^ + | energy_dist1A.inp ​ | 1 | ... | + | energy_dist2A.inp ​ | 2 | ... | + | energy_dist3A.inp ​ | 3 | ... | + | ... | ... | ... | + + This is the Lennard Jones energy curve for two Kr atoms. + By using any plotting program you can now get a representation of the energy profile. + + === 3. Step === + Here are reported the LJ parameters for Ne atoms. Those are to replace the Kr parameters in the input file, along with your Ne coordinates that have to replace the Kr coordinates. A new LJ curve for Ne atoms can be now generated. + + <​code>​ + &​NONBONDED ​ + &​LENNARD-JONES ! Lennard-Jones Ne parameters + atoms Ne Ne + ​EPSILON ​ [K_e] 36.831 ​ + SIGMA [angstrom] ​ 2.775 + ​RCUT ​ [angstrom] 25.0 + &END LENNARD-JONES + &​END NONBONDED + &​CHARGE + ATOM Ne + CHARGE 0.0 + &​END CHARGE + ​ + + === 4. Step === + Here are reported the combination rules for pairs unlike pairs, i.e. for pairs of non identical atoms. \\ + Once generated the ε and σ parameters for the couple Kr/Ne, generate once more the LJ dissociation curve. \\ + Compare the "​mixed"​ curve to the two "​pure"​ curves and report the position and depth of the minimum. \\ + + $$\sigma_{ij}= \sqrt{\sigma_i\sigma_j}$$ \\ + $$\epsilon_{ij}= \sqrt{\epsilon_i\epsilon_j}$$ + + + Remember that you are running ​ with two different atom types. For this reason some of the input sections MUST BE REPLICATED for the two kinds of atoms present + ​ + + * The " LENNARD-JONES " section must be present for all the three possible couples: Kr-Kr, Ne-Ne and Ne-Kr. ​ Example: ​ + + <​code>​ + &​LENNARD-JONES ! Lennard-Jones parameters for Ar-Ar interaction + atoms Kr Kr + EPSILON ​ [K_e] 164.56 + SIGMA [angstrom] ​ 3.601 + RCUT [angstrom] ​ 25.0 + &END LENNARD-JONES + &​LENNARD-JONES ! Lennard-Jones Ne-Ne parameters + atoms Ne Ne + ​EPSILON ​ [K_e] 36.831 ​ + SIGMA [angstrom] ​ 2.775 + ​RCUT ​ [angstrom] 25.0 + &​END LENNARD-JONES + &​LENNARD-JONES ! Lennard-Jones parameters for Kr-Ne interaction + atoms Kr Ne + EPSILON ​ [K_e] YOUR EPSILON + SIGMA [angstrom] ​ YOUR SIGMA + RCUT [angstrom] ​ 25.0 + &END LENNARD-JONES ​ + ​ + + * The " CHARGE " section must be also duplicated: ​ + + <​code>​ + &​CHARGE + ATOM Ne + CHARGE 0.0 + &​END CHARGE + &​CHARGE + ATOM Kr + CHARGE 0.0 + &​END CHARGE + + ​ + + ===== Questions ===== + * Sketch the LJ energy curve for the two set of parameters ($\sigma$and$\epsilon\$) provided. ​ + * Report, for both curves, the minimum energy distance and the depth of the minimum. + * What are the major differences between the curves? How do they relate to the sets of parameters provided?