Why to use a scaling factor for both vdw and electrostatics in QM/MM calculations

[Dawid das]

Yep, so this constitutes to what you wrote in your first message (1/2 =0.5, 
and 1/1.2 = 0.833333)


W dniu środa, 6 czerwca 2018 11:32:14 UTC+2 użytkownik binj... at gmail.com 
napisał:
>
> Now, I got some clue from Amber manual, By default, vdW 1-4 interactions 
> are scaled down by a factor of 2.0, and the electrostatic 1-4 terms by a 
> factor of 1.2. 
>
> 在 2018年6月6日星期三 UTC+8下午5:22:23，binj... at gmail.com写道：
>>
>> Dear Dawid, Thanks for you answer, I am using the standard Amber 
>> forcefield. 
>>
>> best
>>
>> binju 
>>
>> 在 2018年6月6日星期三 UTC+8下午5:15:03，Dawid das写道：
>>
>>> Hello,
>>>
>>> Whether you scale the 1-4 interctions or not depends on the force field 
>>> and system (lipid, protein, nucleic 
>>> acid). For instance, I use CHARMM22 force field and I do not scale the 
>>> 1-4 interactions (VDW_SCALE14 and EI_SCALE14 set to 1.0).
>>>
>>> You need to read how your force field was optimized and what scaling 
>>> factor (if any) is suggested.
>>>
>>> Best wishes,
>>> Dawid
>>>
>>> W dniu środa, 6 czerwca 2018 11:09:41 UTC+2 użytkownik 
>>> binj... at gmail.com napisał:
>>>>
>>>> I saw most of QM/MM input use the scaling factor for both vdw ( 
>>>> VDW_SCALE14 0.5)  and  electrostatics  ( EI_SCALE14 0.8333333).
>>>> It is not clear to me why we need to do this, since we did not us the 
>>>> scaling factor in the classical MD, and the forcefield works well in the 
>>>> classical MD.
>>>> Now, we invoke the same forcefield in QM/MM calculation, but we need to 
>>>> scale the vdw and  electrostatics ?  Thanks!
>>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20180606/7e48258a/attachment.htm>