Understanding the Thole damping effect in MBX 1.3

[Xiaoge Su]

Hello,

I am trying to learn the dipole polarization related topics by reading MBX source code. I am building the MBX using configurations

```bash

./configure --enable-debug --disable-mpi --disable-optimization
```

and running the single point energy on example   examples/PEFs/001_mbpol/C++/ using the following command:

```bash

./single_point input.nrg mbx.json
```

I am having the right energy result, the same to the expected output:

```bash

Energy= 6.4733223872e+00
```

however, when I am trying to understand how the MBX calculates the electrostatic interaction, I have noticed that the function

```cpp

ElectricFieldHolder::CalcPermanentElecField

```

have been used. To the best extent of my understanding, the value `s0r` is used to be the Thole damping value for charge-charge interactions. Yet its value is always zero when treating the three water molecules in the example. The reason is that, the corresponding elec_scaling_factor is always zero when the function is called, which, much to my surprise. This is confirmed by printing out every `s0r` value among all particle pairs. I am not quite sure if this is expected. May I have some help here?

[Ethan Bull-Vulpe]

Hello,

The reason that `s0r` is always 0 when `CalcPermanentElecField` is called for the water trimer is because this function is only called for the intramolecular part of the permanent electric field. Since all the intramolecular interactions are excluded for a small monomer such as water, the `elec_scaling_factor` is always 0, so `s0r` is always 0.

For the intermolecular interactions, a more efficient version of the function `CalcPermanentElecField_Optimized` is called instead (see just below `CalcPermanentElecField` in the same file). If you print out `s0r` in this function, you should see non-zero results.