1. Double-click the Materials Science icon
   • (No icon? See Starting Maestro)

Figure 2-1. Change Working Directory option.

2. Go to File > Change Working Directory
3. Find your directory, and click Choose
4. Pre-generated files are included for running jobs or examining output. Download the zip file here: schrodinger.com/sites/default/files/s3/release/current/Tutorials/zip/polymer_property_prediction.zip
5. After downloading the zip file, unzip the contents in your Working Directorythe location where files are saved for ease of access throughout the tutorial

Figure 2-2. Save Project panel.

6. Go to File > Save Project As
7. Change the File name to polymer_property_prediction_tutorial, click Save
   • The project is now named polymer_property_prediction_tutorial.prj

Figure 2-3. Importing the crosslinked polymers.

First, let’s import five configurations of the equilibrated crosslinked TGDDM-3,3-DDS system to analyze its properties. These five crosslinked polymer systems were prepared as detailed in the Crosslinking Polymers tutorial. For practice building crosslinked polymer systems, revisit that tutorial

 

8. Go to File > Import Structures
9. Navigate to where you downloaded the tutorial files (presumably your working directory) and choose disordered_system_TGDDM_33DDS_n_system-out.cms from the provided files, where n = 1, 2, 3, 4 and 5
10. Click Open
    • Five new entry groups are added to the entry lista simplified view of the Project Table that allows you to perform basic operations such as selection and inclusion containing entries titled disordered_system_TGDDM_33DDS_n

Figure 2-4. Entry list and workspace after importing.

Feel free to visualize and stylize the imported structures

Figure 3-1. The crosslinked systems selected and the Thermophysical Properties panel open.

We will run the calculations on all five equilibrated systems to have a more accurate prediction of T_g and CTE. Best practice is to use ten replicates in these types of calculations, however five is sufficient in this example.

 

1. Select all five crosslinked systems
2. Go to Tasks > Materials > Classical Mechanics > Thermophysical Properties > Thermophysical Property Calculations
   • The Thermophysical Properties panel opens

Figure 3-2. Setting the Thermophysical Property parameters.

To compute T_g and CTE, a temperature range that spans both the glassy (low temperature) and rubbery (high temperature) regions of the system needs to be defined. We will select a temperature range from 100 to 800 K for this example. 

 

3. In the Compute density section, for Over a range of temperatures between, enter 100 K and 800 K in steps of 10 K
   • This will generate a total of 71 data points
4. Ensure Start each temperature with the converged system from the previous temperature is checked
   • We can run the calculation from “High to Low” or “Low to high” temperature, here we leave the default of High to low
   • The simulations start at 800 K and cools down to 100 K with each previously converged system
5. Set Simulation time to 10 ns
   • At each temperature, the density convergence will be checked after the simulation time, 4 ns
6. Set Maximum cycles to 3
   • If the density is not converged after the simulation time, more MD simulations of the same time length will be performed if the Maximum cycles parameter is set to a number greater than 1
7. Ensure Convergence is kept at 5%
8. Maintain Isotropic for Barostat isotropy
   • Here we will study the thermophysical properties of the system isotropically. If interested in how the linear thermal coefficient of expansion changes with direction, the Anisotropic option can be used
9. Check Relax model system before simulation
10. Set the random number seed to 1229

Figure 3-3. Running the Thermophysical Property job.

11. Change the Job name to thermophysical_prop_TGDDM_33DDS
12. Adjust the job settings () as needed
    • This job requires a GPU host. The job can be completed in 10 hours on a GPU host
13. If you would like to run the job yourself, click Run. Otherwise, go to File > Import Structures, navigate to the provided tutorial files and Open:  Section_03 > thermophysical_prop_TGDDM_33DDS > thermophysical_prop_TGDDM_33DDS-00n-out.cms where n = 1-5
14. Close the Thermophysical Properties panel

 

MD Simulations have a number of files associated with the job, for a full description of each file type see the help documentation on Desmond Files.

Figure 3-4. Viewing the output system.

15. When the job is finished or after importing, select(1) the atoms are chosen in the Workspace. These atoms are referred to as "the selection" or "the atom selection". Workspace operations are performed on the selected atoms. (2) The entry is chosen in the Entry List (and Project Table) and the row for the entry is highlighted. Project operations are performed on all selected entries and includethe entry is represented in the Workspace, the circle in the In column is blue one of the new 100.0_thermophysical_prop_TGDDM_33DDS-00n-out entries from the corresponding MD: disordered_system_TGDDM_33DDS_n_system (1) entry group
    • The displayed cell is from the last simulation at 100 K. If many atoms end up outside the unit cell; this is purely visual, but can be easily amended  using the Periodic Structure Tool Window
    • Feel free to visualize the output system in the workspacethe 3D display area in the center of the main window, where molecular structures are displayed

Figure 3-5. Opening the Thermophysical Properties Results panel via the WAM.

To get T_g and CTE from the Thermophysical Properties calculation we use the Thermophysical Properties Results panel:

 

16. Select(1) the atoms are chosen in the Workspace. These atoms are referred to as "the selection" or "the atom selection". Workspace operations are performed on the selected atoms. (2) The entry is chosen in the Entry List (and Project Table) and the row for the entry is highlighted. Project operations are performed on all selected entries and includethe entry is represented in the Workspace, the circle in the In column is blue 100.0_thermophysical_prop_TGDDM_33DDS-003-out
17. Use the WAM (workflow action menu) button () to open the Thermophysical Properties Results panel
    • Alternatively, access the panel via Tasks > Materials > Classical Mechanics > Thermophysical Properties Results
    • The Thermophysical Properties Results panel opens

Figure 3-6. Viewing the Hyperbola fit in the Thermophysical Properties Results panel.

Two methods are available to obtain the thermophysical properties of the system: a Bilinear fit, in which a linear regression is performed on the low temperature (glassy) region and the high temperature (rubbery) region of the plot of density versus temperature; and a Hyperbola fit, in which a single hyperbolic curve is fitted to all points of the plot of density versus temperature. By default, the Hyperbola fit is selected.

 

For replicate 3, the T_g is ~ 539 K. For the other replicates this value may be slightly different because of variability in the densities calculated. We can generally reduce the scatter or variability in the density data with larger systems and longer MD simulation times. We will learn more about the spread in T_g over all the replicates after calculating CTE for this replicate.

 

Now, let’s see what a Bilinear fit would give for T_g and more importantly, CTE:

 

18. Select Bilinear fit

Figure 3-7. Viewing the Bilinear fit in the Thermophysical Properties Results panel.

19. Click and move the edges of the blue and green boxes to readjust the boundaries between the low temperature (green) and high temperature (blue) regions to shift to the most linear regions. In the Figure, the green and blue regions are approximately set to 100 K to 460 K and 585 K to 800 K, respectively.
    • The resulting T_g value should be ~528 K which is slightly different than that given by the hyperbolic fit
    • The coefficient of thermal expansion is determined from the slope of the fit in the glassy region. Here, the volumetric and linear CTE are reported as ~107 x 10^-6 K^-1 and 36 x 10^-6 K^-1 respectively
20. Close the Thermophysical Properties Results panel

 

Note: The Store CTE and Tg button allows you to save the properties when you are satisfied with the fit to the data. The property is saved to the includedthe entry is represented in the Workspace, the circle in the In column is blue structure and can be displayed in the Project Tabledisplays the contents of a project and is also an interface for performing operations on selected entries, viewing properties, and organizing structures and data

Figure 3-8. Opening the Uncertainty Quantification panel.

We have generated thermophysical data for five replicates of the system to ensure we have reliable data. Instead of viewing the T_g for each replicate using the methods described above, the Uncertainty Quantification panel can be used to efficiently calculate the spread of the thermophysical quantities.

 

21. Select(1) the atoms are chosen in the Workspace. These atoms are referred to as "the selection" or "the atom selection". Workspace operations are performed on the selected atoms. (2) The entry is chosen in the Entry List (and Project Table) and the row for the entry is highlighted. Project operations are performed on all selected entries the output of all five thermophysical property calculations, 100.0_thermophysical_prop_TGDDM_33DDS-00n-out
22. Go to Tasks > Materials > Tools > Uncertainty Quantification Calculations
    • The Uncertainty Quantification panel opens

Figure 3-9. Setting the Uncertainty Quantification parameters.

This panel is used to set up a job for the uncertainty quantification of the prediction of T_g or the yield strain from multiple thermophysical property simulations. Here, we will maintain the defaults as we are interested in analyzing the uncertainty in our five Tg calculations based on a Hyperbola fit

 

Pooling Analysis gives the average deviation when we take different combinations of replicates (pools). As we increase the pool size, the average T_g value should converge and the standard deviation should decrease. If we see convergence, it indicates that the number of replicates used is sufficient.

 

23. Click Add Row four times
    • Four new rows are added to the Pooling Analysis table
24. In the four new rows, add the following pairs of Pool Size and Number of Pools: (2, 10), (3, 10), (4, 5) and (5,1)
    • By selecting all possible pool sizes, we can check the dependence of the results on the system size
25. Check Skip failed fits
    • In case any hyperbola fitting fails, the analysis will still run.

Figure 3-10. Running the Uncertainty Quantification calculation. Note that Estimated Pooling Time may vary slightly.

26. Change the Job name to uncertainty_quantification_TGDDM_33DDS
27. Adjust the job settings () as needed
    • This job requires a CPU host and can be completed in a few minutes
28. If you would like to run the job yourself, click Run. Otherwise, go to File > Import Structures, navigate to the provided tutorial files and Open:  Section_03 > uncertainty_quantification_TGDDM_33DDS > uncertainty_quantification_TGDDM_33DDS_out.mae
29. Close the Uncertainty Quantification panel

Figure 3-11. Visualizing the output.

30. When the job is finished or after importing, select(1) the atoms are chosen in the Workspace. These atoms are referred to as "the selection" or "the atom selection". Workspace operations are performed on the selected atoms. (2) The entry is chosen in the Entry List (and Project Table) and the row for the entry is highlighted. Project operations are performed on all selected entries and includethe entry is represented in the Workspace, the circle in the In column is blue the new 100.0_thermophysical_prop_TGDDM_33DDS-001-out entry from the Uncertainty Quantification Calculation
    • The one representative structure now contains T_g information for all five replicates

Figure 3-12. Opening the Uncertainty Quantification Results panel via the WAM.

We can analyze the Uncertainty Quantification calculation further using the Uncertainty Quantification Results panel:

 

31. Select(1) the atoms are chosen in the Workspace. These atoms are referred to as "the selection" or "the atom selection". Workspace operations are performed on the selected atoms. (2) The entry is chosen in the Entry List (and Project Table) and the row for the entry is highlighted. Project operations are performed on all selected entries and includethe entry is represented in the Workspace, the circle in the In column is blue 100.0_thermophysical_prop_TGDDM_33DDS-001-out
32. Use the WAM button () to open the Uncertainty Quantification Results panel
    • Alternatively, access the panel via Tasks > Materials > Tools > Uncertainty Quantification Results
    • The Uncertainty Quantification Results panel opens

Figure 3-13. Viewing the Aggregate Analysis results.

In the Aggregate tab, plots of T_g with uncertainty bars are displayed. At the bottom of the panel, you can find information on the aggregate and arithmetic averages and standard deviations for the five replicates of the T_g calculation. The aggregate average, 505.46 K, is a weighted average while the arithmetic average, 530.39 K, is the simple mean.

 

33. Go to the Pooling Analysis tab

Figure 3-14. Viewing the Pooling Analysis results.

The Pooling Analysis tab displays plots based on pooling information to check the dependence of the results on the system size. This shows that increasing the number of replicates for the T_g calculation allows us to have a better computational prediction of what the T_g of the system will be. The pooled T_g for a large enough dataset should not deviate much as the pool size increases. If a large change is seen as the pool size increases, then the number of replicates may not be sufficient

 

34. Close the Uncertainty Quantification Calculation panel

Figure 4-1. The crosslinked systems selected and the Stress Strain panel open.

We will run stress-strain calculations on all five equilibrated crosslinked systems imported in Section 2. Best practice is to use ten replicates in these types of calculations, however five is sufficient in this example.

 

1. Select all five crosslinked systems imported in Section 2
   • The MD system for Stress Strain calculations could also be prepared using Machine Learning Force Field (MLFF) methods. For further details, please refer to the Machine Learning Force Field tutorial.
2. Go to Tasks > Materials > Classical Mechanics > Stress Strain > Stress Strain Calculations
   • The Stress Strain panel opens

Figure 4-2. Setting the Stress Strain parameters.

3. For Type, select Volume conserving uniaxial from the drop-down menu

 

In this example, the Main axis is left as the default “a”. Ideally, this calculation should deform all three axes

 

4. For Number of steps, enter 100
   • There will be 100 steps of increasing strain along the main axis
5. For Step size, enter 0.002
   • The step size gives the size of the strain increment along the main axis
6. For Simulation protocols, set Simulation time to 20.0 ps
   • At each value of strain, a 20 ps MD simulation will run
7. Set Trajectory recording interval to 1.0 ps
8. Ensure Temperature is set to 300K
9. Check Combined trajectory

For a more in depth understanding of the Stress Strain panel and the parameters shown here, please visit the help documentation

Figure 4-3. Running the Stress Strain calculation.

10. Change the Job name to stress_strain_TGDDM_33DDS
11. Adjust the job settings () as needed
12. This job requires a GPU host. The job can be completed in 2 hours on a GPU host
13. If you would like to run the job yourself, click Run. Otherwise, go to File > Import Structures, navigate to the provided tutorial files and Open  Section_04 > stress_strain_TGDDM_33DDS-00n > stress_strain_TGDDM_33DDS-00n-out.cms
14. Close the Stress Strain panel

Figure 4-4. Viewing the output of the Stress Strain simulation.

15. When the job is finished or after importing, select(1) the atoms are chosen in the Workspace. These atoms are referred to as "the selection" or "the atom selection". Workspace operations are performed on the selected atoms. (2) The entry is chosen in the Entry List (and Project Table) and the row for the entry is highlighted. Project operations are performed on all selected entries and includethe entry is represented in the Workspace, the circle in the In column is blue the new disordered_system_TGDDM_33DDS_3 entry from the MD: disordered_system_TGDDM_33DDS_3_system entry group
    • Feel free to visualize the output system in the workspace

Figure 4-5. Opening the Stress Strain Results panel via the WAM.

We can analyze the Stress Strain calculation further using the Stress Strain Results panel:

 

16. Use the WAM button () to open the Stress Strain Result panel
    • Alternatively, access the panel via Tasks > Materials > Classical Mechanics > Stress Strain > Stress Strain Results
    • The Stress Strain Results panel opens

Figure 4-6. Viewing the stress versus strain plot.

The Stress Strain Results panel opens with a plot of the effective stress against the effective strain from the calculation we just ran. The data shown here can be used to obtain the yield point of the crosslinked TGDDM-3,3-DDS system, however there is some noise in the data. It is much more effective to load in all five stress strain simulations, and average the stress strain data over the replicates. We will do so in the next step.

Figure 4-7. Load data from multiple replicates.

17. Back in the entry lista simplified view of the Project Table that allows you to perform basic operations such as selection and inclusion, select all five outputs of the Stress Strain calculations, disordered_system_TGDDM_33DDS_all_components_amorphous_n
18. In the Stress Strain Results panel, click Load Averaged Data from Selected Entries
    • Aggregated results from all of our simulations on the system are shown. The shaded area around the plot line indicates the standard deviation over the multiple replicates

Figure 4-8. Averaged data from multiple replicates.

19. For Stress type, choose Normal from the drop down menu
    • The Strain component automatically updates to Normal
    • Normal stress is the raw stress value from the simulation. For more on stress and strain types, visit the thorough help documentation
20. Go to the Convex analysis tab

Figure 4-9. The Convex analysis.

We will now use Convex analysis to find the yield point. The plot on the left is the same as that in the General tab while the plot on the right will show the strain at maximum stress as a function of the strain used in the fit. We expect the curve to be linear at first and then reach a plateau. The value of the strain at the plateau is the predicted yield strain 

 

21. Click on Compute
    • The plot on the right is populated

 

By using the cursor to trace the plateau we determine the yield strain is approximately 0.083. We can then determine the normal stress at that value of strain in the left-hand plot. In doing so, we see the yield point of this system is 0.32 GPa    

Figure 4-10. Selecting the five outputs of the Stress Strain simulation and opening the Free Volume Calculations panel.

Free Volume calculations can be performed on the output of our Stress Strain calculations to determine the size and location of the voids in the new strained system. The location and sizes of the voids in the structures are determined by a grid-based method. Analyzing void nucleation can be helpful in more complex polymeric systems involving other components, such as interfaces. We will use the defaults in this example as it is sufficient for our calculation, but feel free to experiment with the grid spacings as you would like. For more details on this panel, visit the Free Volume tutorial.

 

22. Select all five outputs from the stress strain calculations
23. Go to Tasks > Materials > Tools > Free Volume Calculations
    • The Free Volume Analysis panel opens

Figure 4-11. Setting the Free Volume parameters.

24. Ensure Use structures from shows Project Table (5 selected entries)
25. Check the Analyze trajectory check box
26. Click Trajectory Frames…

Figure 4-12. Setting the Trajectory Frames.

27. Set the range to 1894 to 1919
    • We will only run the Free Volume analysis on the last 26 frames of the stress strain simulation
28. Click OK

Figure 4-13. Running the Free Volume calculation.

29. Change the Job name to free_volume_TGDDM_33DDS
30. Adjust the job settings () as needed
    • This job requires a CPU host. The job can be completed in 5 hours on a single CPU host
    • If you have multiple CPUs, you can run the job on each of the five structures individually. The five calculations would have to be submitted manually, but the set of calculations would finish quicker
31. If you would like to run the job yourself, click Run. Otherwise, go to File > Import Structures, navigate to the provided tutorial files and Open: Section_04 > free_volume_TGDDM_33DDS >  disordered_system_TGDDM_33DDS_3-27-out.cms
    • Please note that disordered_system_TGDDM_33DDS_n-27-out.cms for n = 1, 2, 4, 5 can also be imported, but we will only visualize n = 3 here
32. Close the Free Volume panel

Figure 4-14. Visualizing the output of the Free Volume calculation.

33. When the job is finished or after importing, select(1) the atoms are chosen in the Workspace. These atoms are referred to as "the selection" or "the atom selection". Workspace operations are performed on the selected atoms. (2) The entry is chosen in the Entry List (and Project Table) and the row for the entry is highlighted. Project operations are performed on all selected entries and includethe entry is represented in the Workspace, the circle in the In column is blue the new disordered_system_TGDDM_33DDS_3 entry from the MD: disordered_system_TGDDM_33DDS_3_system entry group
    • Feel free to visualize the output system in the workspacethe 3D display area in the center of the main window, where molecular structures are displayed
    • It is possible for the unit cell and crosslinked polymer to become misaligned with the periodic structure; this is purely visual, but can be easily amended in the next step

Figure 4-15. Aligning the unit cell.

We can fix the visualization of the system using the following steps:

34. Click on the button to open the Periodic Structure Tool Window
35. Go to Build Cell
36. Click Translate to First Unit Cell
37. Click Apply
    • Now, all of the atoms comprising one unit cell are shown and are translated to the unit cell represented by the blue box in the workspacethe 3D display area in the center of the main window, where molecular structures are displayed

 

The steps we just performed are purely visual and do not change anything about the structure or its properties

Figure 4-16. Opening the Free Volume Results panel via the WAM.

We can visualize the results of the Free Volume calculation using the Free Volume Results panel:

 

38. Use the WAM button () to open the Free Volume Results panel
    • Alternatively, access the panel via Tasks > Materials > Tools > Free Volume Results
    • The Free Volume Results panel opens

Figure 4-17. Viewing the Distribution tab.

39. Click Load Data From Workspace
    • The results from the Free Volume calculation are loaded into the panel.
40. Back in the workspacethe 3D display area in the center of the main window, where molecular structures are displayed, change the Current frame to 26
    • We will look at the voids in the last frame of the stress strain simulation

 

The Void Size Distribution plot shows the distribution of voids in this particular frame.

41. Go back to the Free Volume Results Viewer and change the lower radii limit for Show voids in Workspace for current frame with radii between to 1.31
    • The Number of voids in current frame updates to 2
42. Click Display to view the two largest voids in this frame
    • Feel free to visualize a greater number of voids but be aware that the voids will take some time to load into the workspacethe 3D display area in the center of the main window, where molecular structures are displayed
43. Go to the Location tab

Figure 4-18. Viewing the Location tab.

The Location tab displays the location of the voids in a cross-section of the Workspacethe 3D display area in the center of the main window, where molecular structures are displayed frame, taken perpendicular to a coordinate axis. A plane is shown in the Workspacethe 3D display area in the center of the main window, where molecular structures are displayed that indicates the location of the current cross-section. Feel free to explore the different cross-sections as interested

 

44. Go to the Convergence tab

Figure 4-19. Viewing the Convergence tab.

The Convergence tab displays a plot of the convergence of the free volume as a function of the grid spacing. Here we see that the convergence of the free volume is not dependent on the grid spacing

 

45. Close the Free Volume Analysis Viewer

 

Feel free to further explore the Free Volume visualizations if you wish

Figure 4-20. Viewing the voids in the initial crosslinked configurations.

An interesting exercise is to compare the voids in the strained structure to the unstrained structure. By following the steps above for the Free Volume calculations and subsequent analysis, similar results can be generated for the five initial structures we began the tutorial with. Provided in the tutorial files is the free_volume_initial folder which contains the outputs of a free volume calculation (Section_04 > free_volume_initial >

free_volume_initial-freevolume.maegz) on the initial configurations. Feel free to compare the voids before and after the stress strain simulation. 

Figure 5-1. Opening the Elastic Constants calculation panel.

We will run the elastic constants calculation on all five equilibrated crosslinked systems imported in Section 2.

 

1. Select all five crosslinked systems imported in Section 2
   • The MD system for Elastic Constants calculations could also be prepared using Machine Learning Force Field (MLFF) methods. For further details, please refer to the Machine Learning Force Field tutorial.
2. Go to Tasks > Materials > Classical Mechanics > Elastic Constants > Elastic Constants Calculations
   • The Elastic Constants panel opens

Figure 5-2. Setting the parameters for the Elastic Constants calculation.

3. Ensure Use structures from shows Project Table (5 selected entries)
4. For Energy group recording interval, enter 0.01

 

We will retain the defaults for the rest of the settings. The Uniaxial and Biaxial tension will be computed with 8 increments in 0.002 increments. The total amount of strain (n times ε) should be small to ensure the material is in the elastic region (for example < 0.02 or 2 % strain). The n should be set so there are enough points to obtain a reliable linear fit ( generally > 4). The defaults work well for most polymeric materials.

Figure 5-3. Running the Elastic Constants job.

5. Change the Job name to elastic_constants_stress_based
6. Adjust the job settings () as needed
7. This job requires a GPU host. The job can be completed in 5.5 hours on a GPU host when using 2 GPUs per structure
8. If you would like to run the job yourself, click Run. Otherwise, go to File > Import Structures, navigate to the provided tutorial files and Open  Section_05 > elastic_constants_stress_based > elastic_constants_stress_based-00n > elastic_constants_stress_based-00n-out.cms
9. Close the Elastic Constants panel

Figure 5-4. Viewing the output.

10. When the job is finished or after importing, select(1) the atoms are chosen in the Workspace. These atoms are referred to as "the selection" or "the atom selection". Workspace operations are performed on the selected atoms. (2) The entry is chosen in the Entry List (and Project Table) and the row for the entry is highlighted. Project operations are performed on all selected entries and includethe entry is represented in the Workspace, the circle in the In column is blue the new disordered_system_TGDDM_33DDS_3 entry from the MD: disordered_system_TGDDM_33DDS_3_system entry group
    • Feel free to visualize the output system in the workspace

Figure 5-5. Opening the Elastic Constants Viewer panel via the WAM.

We can analyze the Elastic Constants calculations further using the Elastic Constant Results panel:

 

11. Use the WAM button () to open the Elastic Constants Results panel
    • Alternatively, access the panel via Tasks > Materials > Classical Mechanics > Elastic Constants > Elastic Constants Results
    • The Elastic Constants Results panel opens

Figure 5-6. The Plot tab of the Elastic Constants Results panel.

For the results of an Elastic Constants calculation, the Plot tab shows the stress computed against the deformation along with the linear fit. The Term can be changed to see the stress/deformation curve for any of the tensor components, C_ij. Feel free to explore these plots.

 

12. Go to the Tensors tab

Figure 5-7. The Tensors tab.

The Tensors tab displays the Elastic tensor table, containing the components of the elastic tensor, and the R² table, which contains the R² value for the fit to obtain each component of the elastic tensor.

 

Feel free to further explore the polymer properties of interest