CP2K Open Source Molecular Dynamics

User Tools

Site Tools

Trace:
exercises:2021_uzh_acpc2:ex03

Differences

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

Link to this comparison view

Both sides previous revisionPrevious revision
Last revisionBoth sides next revision
exercises:2021_uzh_acpc2:ex03 [2021/05/21 07:50] – [NEB activation barrier] Some improvements mrossmannekexercises:2021_uzh_acpc2:ex03 [2021/05/21 08:09] – [Free energy surface] Some improvements mrossmannek
Line 169: Line 169:
 Sampling the free energy surface (FES) of a chemical system is a convenient method to explore various stable conformations and possible reaction pathways. To calculate the FES for complicated systems, advanced sampling methods (such as umbrella sampling, metadynamics, parallel tempering, ...) have to be used. However, for our simple $S_N2$ reaction, we will use unbiased Molecular Dynamics (MD).  Sampling the free energy surface (FES) of a chemical system is a convenient method to explore various stable conformations and possible reaction pathways. To calculate the FES for complicated systems, advanced sampling methods (such as umbrella sampling, metadynamics, parallel tempering, ...) have to be used. However, for our simple $S_N2$ reaction, we will use unbiased Molecular Dynamics (MD). 
  
-The FES is a projection of the high-dimensional free energy landscape into a small number, usually two, dimensions. These two dimensions are called collective variables (CV) and they must be chosen such that various stable conformations can be distinguished and the reaction pathways can be adequately described by the FES. For complex systems, the choice of CVs is a non-trivial task. Fortunately for our simple system, the choice is simple: we take the distances of the two Cl anions from the central C as the CVs.+The FES is a projection of the high-dimensional free energy landscape onto a small number, usually two, dimensions. These two dimensions are called collective variables (CV) and they must be chosen such that various stable conformations can be distinguished and the reaction pathways can be adequately described by the FES. For complex systems, the choice of CVs is a non-trivial task. Fortunately, for our simple system, the choice is simple: we take the distances of the two ''Cl'' anions from the central ''C'' as the CVs.
  
-To help the calculation sample the parts of FES we care about, we will include restraints in the MD run, which prevent the Cl anions from going too far from the molecule.+To help the simulation sample the parts of the FES which we care about, we will include restraints in the MD run, which prevent the ''Cl'' anions from going too far away from the molecule.
  
 The following CP2K input script runs our MD calculation and prints out the CV values for every step: The following CP2K input script runs our MD calculation and prints out the CV values for every step:
Line 274: Line 274:
 $$ F(s) = -k T \log(P(s)),$$ $$ F(s) = -k T \log(P(s)),$$
  
-where $s$ is the set CVs and $P(s)$ is the probability that the system has the set of CV values $s$.+where $s$ is the set of CVs and $P(s)$ is the probability that the system has the set of CV values $s$.
  
-The following Python script can be used to calculate the FES from the ''ch3f-COLVAR.metadynLog'' file produced by the MD run. (Don't forget to change the temperature in Python script!)+The following Python script can be used to calculate the FES from the ''ch3f-COLVAR.metadynLog'' file produced by the MD run. (**Don't forget to change the temperature in the Python script!**)
  
 <code> <code>
Line 285: Line 285:
 kb = 8.6173303e-5 # eV * K^-1 kb = 8.6173303e-5 # eV * K^-1
  
-temperature = 1000.0                          #Change temperature according to your MD simulations!+temperature = 1000.0                          # Change the temperature according to your MD simulations!
 colvar_path = "./ch3f-COLVAR.metadynLog" colvar_path = "./ch3f-COLVAR.metadynLog"
  
Line 316: Line 316:
 </code> </code>
  
-Here is an example output for temperature 1000K. We clearly see the two local minima corresponding to the one of the chlorine atoms being covalently bonded (distance 1.8 Å) while the other one is around distance 2.5 Å.+Here is an example output for a temperature of 1000K. We clearly see the two local minima corresponding to one of the chlorine atoms being covalently bonded (distance 1.8 Å) while the other one is around a distance of 2.5 Å.
  
 {{  :exercises:2019_uzh_acpc2:fes1000.png ?direct&500 |}} {{  :exercises:2019_uzh_acpc2:fes1000.png ?direct&500 |}}
  
 <note>**TASK 3** <note>**TASK 3**
-  * Run the MD calculation for 400K, 800K, 1200K and 1600K. (The calculations can a take a while.)+  * Run the MD calculation for 400K, 800K, 1200K and 1600K. (These calculations can a take a while.)
   * Create the corresponding FES plots and discuss the temperature dependence.   * Create the corresponding FES plots and discuss the temperature dependence.
   * In general, how does potential energy differ from free energy? For our reaction, what are the activation barriers from the different free energy surfaces? How and why do they differ from the NEB barrier?   * In general, how does potential energy differ from free energy? For our reaction, what are the activation barriers from the different free energy surfaces? How and why do they differ from the NEB barrier?
 </note> </note>
exercises/2021_uzh_acpc2/ex03.txt · Last modified: 2021/05/21 10:13 by mrossmannek