van der Waals cutoff vs. system size

Skip to first unread message

Binquan Luan

Nov 24, 2021, 1:43:37 PM11/24/21
to cp2k
Dear all,
For DFT calculations with van der Waals corrections, I am wondering what the right cutoff distance should be. In the manual, I found


Range of potential. The cutoff will be 2 times this value  [Edit on GitHub]

This keyword cannot be repeated and it expects precisely one real.

Default value: 1.05835442E+001

Default unit: [angstrom]

If the default range of potential is 21 angstrom, with the minimum image convention used for PBC, the system size (in each dimension) should be at least 42 angstrom that is much larger than most systems reported in literature.  Do I miss anything here?

Thanks a lot for help!


Nov 25, 2021, 3:19:40 AM11/25/21

Minimum image convention is not applied in PBC calculations in CP2K.


Juerg Hutter
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie E-mail:
Universität Zürich
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
--------------------------------------------------------------- wrote: -----
To: "cp2k" <>
From: "Binquan Luan"
Sent by:
Date: 11/24/2021 07:43PM
Subject: [CP2K:16303] van der Waals cutoff vs. system size
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
To view this discussion on the web visit

Binquan Luan

Nov 25, 2021, 9:30:08 AM11/25/21
to cp2k
Hi Juerg,
Thanks for clarification.  Can you help explain how the van der Waals interaction is calculated when the cutoff distance is much larger than the primitive cell. One example is a small molecule on the material surface. To calculate the interaction between a small molecule and the substrate surface, is the substrate surface duplicated (with the PBC)?
Reply all
Reply to author
0 new messages