QM region heating up and crashing

[Dries Van Rompaey]

Hi everyone,

I'm running a simulation of the Chorismate Mutase complex in water with 
CP2K. The enzyme and waters are treated at the MM level, while one copy of 
the chorismate is treated at the QM level (PM6). When I run the system with 
ECOUPL NONE, everything works fine. However, when I try to switch to 
Coulomb coupling, the system crashes at around 950 steps. The QM temperate 
skyrockets and I get the error:

Particles:    5766   4633 at distance [au]:     0.79662080 less than:     
>  0.95400000; increase EMAX_SPLINE. 
> GEOMETRY wrong or EMAX_SPLINE too small 


 I suspect that the error is due to the interaction of the tip3p waters 
with the negatively charged QM region. I tried to fix this by giving them a 
radius through MM_KIND RADIUS, but that unfortunately did not work. I have 
also tried adding LJ-interactions between the negative carboxylate and the 
water hydrogens (currently commented out in my input file) to no success. 

I have uploaded input and output files to dropbox  (
https://www.dropbox.com/sh/du6f32omx32d7a5/AAAiYYttGmwrDj5V_v3y3QZKa?dl=0) 
as they are too big for the forum. I would greatly appreciate some help on 
what causes this error and how to fix it. Please let me know if you require 
any additional information to help troubleshoot this issue. 

Kind regards

Dries
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170614/9bbec0b9/attachment.htm>