Simple molecular dynamics - Revision history

Miguel Caro: /* Melting and expanding */

2021-07-30T14:07:28Z

Melting and expanding

Miguel Caro: /* Melting and expanding */

2021-07-30T11:50:27Z

Melting and expanding

Miguel Caro at 11:49, 30 July 2021

2021-07-30T11:49:11Z

Miguel Caro: /* Melting and expanding */

2021-07-30T11:47:35Z

Melting and expanding

Miguel Caro: /* Melting and expanding */

2021-07-30T11:43:09Z

Melting and expanding

Miguel Caro at 11:42, 30 July 2021

2021-07-30T11:42:30Z

Miguel Caro: /* Graphitization */

2021-05-04T18:39:50Z

Graphitization

Miguel Caro: /* Graphitization */

2021-05-04T18:37:17Z

Graphitization

Miguel Caro: /* Graphitization */

2021-05-03T14:11:49Z

Graphitization

Miguel Caro: /* Melting and expanding */

2021-04-25T17:44:12Z

Melting and expanding

@@ Line 54: / Line 54: @@
 Now you have a <code>gap_files/</code> directory with all the necessary files to run a GAP simulation with '''TurboGAP'''.
-We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Thermostat#Berendsen thermostat|Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following [[Documentation#Files|input file]]:
+We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Thermostat#Berendsen thermostat|Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following [[input file]]:
   ! Species-specific info

@@ Line 54: / Line 54: @@
 Now you have a <code>gap_files/</code> directory with all the necessary files to run a GAP simulation with '''TurboGAP'''.
-We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Thermostat#Berendsen thermostat|Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following [[Documentation#Input file(input)|input file]]:
+We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Thermostat#Berendsen thermostat|Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following [[Documentation#Files|input file]]:
   ! Species-specific info

@@ Line 54: / Line 54: @@
 Now you have a <code>gap_files/</code> directory with all the necessary files to run a GAP simulation with '''TurboGAP'''.
-We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Thermostat#Berendsen thermostat|Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following [[input file]]:
+We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Thermostat#Berendsen thermostat|Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following [[Documentation#Input file(input)|input file]]:
   ! Species-specific info

@@ Line 54: / Line 54: @@
 Now you have a <code>gap_files/</code> directory with all the necessary files to run a GAP simulation with '''TurboGAP'''.
-We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Thermostat#Berendsen thermostat|Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following <code>[[input]]</code> file:
+We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Thermostat#Berendsen thermostat|Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following [[input file]]:
   ! Species-specific info

@@ Line 54: / Line 54: @@
 Now you have a <code>gap_files/</code> directory with all the necessary files to run a GAP simulation with '''TurboGAP'''.
-We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following <code>[[input]]</code> file:
+We are going to melt the diamond structure at 9000 K for 3 ps. We choose the [[Thermostat#Berendsen thermostat|Berendsen thermostat]] with time step 1 fs and time constant 100 fs. At the same time, we are going to expand each of the lattice vectors by a factor of 1.2, since the density of diamond is too high to generate graphitic carbon. You need the following <code>[[input]]</code> file:
   ! Species-specific info

← Older revision		Revision as of 11:42, 30 July 2021
Line 1:		Line 1:
−	In this tutorial we will run a very simple MD simulation with '''TurboGAP'''. We will melt a diamond lattice and then quench the liquid down to 3500K and bake it at that temperature for some time to make graphitic carbon. In this tutorial we will not worry about pressure, but will use the <code>[[~~rescale_box~~]]</code> functionality.	+	In this tutorial we will run a very simple MD simulation with '''TurboGAP'''. We will melt a diamond lattice and then quench the liquid down to 3500K and bake it at that temperature for some time to make graphitic carbon. In this tutorial we will not worry about pressure, but will use the <code>[[scale_box]]</code> functionality.

	== Prerequisites for this tutorial ==		== Prerequisites for this tutorial ==

← Older revision		Revision as of 18:39, 4 May 2021
Line 136:		Line 136:

	Additionally, you can choose to quench the structure to, e.g., room temperature. In that case, just extract the last snapshot from the graphitization run, and use it as starting point for another MD run with both <code>t_beg</code> and <code>t_end</code> set to 300, and reduce the number of steps to 10000 or so.		Additionally, you can choose to quench the structure to, e.g., room temperature. In that case, just extract the last snapshot from the graphitization run, and use it as starting point for another MD run with both <code>t_beg</code> and <code>t_end</code> set to 300, and reduce the number of steps to 10000 or so.
		+
		+	Here you can see the graphitization process and resulting structure:

	<youtube>jft8n10Dg10</youtube>		<youtube>jft8n10Dg10</youtube>

← Older revision		Revision as of 18:37, 4 May 2021
Line 136:		Line 136:

	Additionally, you can choose to quench the structure to, e.g., room temperature. In that case, just extract the last snapshot from the graphitization run, and use it as starting point for another MD run with both <code>t_beg</code> and <code>t_end</code> set to 300, and reduce the number of steps to 10000 or so.		Additionally, you can choose to quench the structure to, e.g., room temperature. In that case, just extract the last snapshot from the graphitization run, and use it as starting point for another MD run with both <code>t_beg</code> and <code>t_end</code> set to 300, and reduce the number of steps to 10000 or so.
		+
		+	<youtube>jft8n10Dg10</youtube>

@@ Line 133: / Line 133: @@
 We can check again the evolution of temperature and potential energy:
-[[File:tutorial_002.png|left|320px|Temperature and potential energy during graphitization simulation.]]
+[[File:tutorial_002.png|center|480px|Temperature and potential energy during graphitization simulation.]]
 Additionally, you can choose to quench the structure to, e.g., room temperature. In that case, just extract the last snapshot from the graphitization run, and use it as starting point for another MD run with both <code>t_beg</code> and <code>t_end</code> set to 300, and reduce the number of steps to 10000 or so.

@@ Line 57: / Line 57: @@
   ! Species-specific info
-  atoms_file = 'diamond.xyz'
+  [[atoms_file]] = 'diamond.xyz'
-  pot_file = 'gap_files/carbon.gap'
+  [[pot_file]] = 'gap_files/carbon.gap'
-  n_species = 1
+  [[n_species]] = 1
-  species = C
+  [[species]] = C
-  masses = 12.01
+  [[masses]] = 12.01
   ! MD options
-  md_nsteps = 3000
+  [[md_nsteps]] = 3000
-  md_step = 1.
+  [[md_step]] = 1.
-  thermostat = berendsen
+  [[thermostat]] = berendsen
-  t_beg = 9000
+  [[t_beg]] = 9000
-  t_end = 9000
+  [[t_end]] = 9000
-  tau_t = 100.
+  [[tau_t]] = 100.
-  write_thermo = 1
+  [[write_thermo]] = 1
-  write_xyz = 10
+  [[write_xyz]] = 10
-  scale_box = .true.
+  [[scale_box]] = .true.
-  box_scaling_factor = 1.2
+  [[box_scaling_factor]] = 1.2
 Then, run '''TurboGAP'''. This MD took 6 minutes on my oldish desktop machine, running on 8 cores, so take this into account. You can either run the code serially: