<div dir="ltr">I am trying to implement q-TIP4P/F (a flexible, 4-point water model: J. Chem. Phys. 131, 024501 (2009)) into cp2k using the VIRTUAL_SITE module to keep the negative charge in place. I was able to implement TIP4P (a rigid model) without any problem, but I get the following error when I try to use q-TIP4P/F: ********************************************************* *** ERROR in get_molecule (MODULE molecule_types_new) *** ********************************************************* *** The pointer lci is not associated *** I assume this is because VIRTUAL_SITE only works with constrained molecules. If so, is there another way to constrain the negative charge for a flexible 4-point model? I’ve inserted the key areas of my input file for TIP4P (which works) and q-TIP4P/F (which doesn’t work) below. I’ve also attached a full input file. Thanks, <div>-- Daniel Sadowsky Ecole Polytechnique Fédérale de Lausanne Environmental Chemistry Modeling Laboratory @IF ( ${MD_FF} == TIP4P ) &CONSTRAINT SHAKE_TOLERANCE 1.0E-12 &G3X3 MOLECULE 2 ATOMS 1 2 3 DISTANCES [angstrom] 0.95720 0.95720 1.51390 &END G3X3 &VIRTUAL_SITE INTERMOLECULAR FALSE ATOMS 4 2 1 3 PARAMETERS 0.128012065 0.128012065 MOLECULE 2 &END VIRTUAL_SITE &END CONSTRAINT @ENDIF @IF ( ${MD_FF} == QTIP4PF ) &CONSTRAINT SHAKE_TOLERANCE 1.0E-12 &G3X3 MOLECULE 2 ATOMS 1 2 3 DISTANCES [angstrom] 0.95720 0.95720 1.51390 &END G3X3 &VIRTUAL_SITE INTERMOLECULAR FALSE ATOMS 4 2 1 3 PARAMETERS 0.131940 0.131940 MOLECULE 2 &END VIRTUAL_SITE &END CONSTRAINT @ENDIF @IF ( ${MD_FF} == QTIP4PF ) &BOND ATOMS Ow Hw KIND QUARTIC K 0.541925 -0.983779 0.926020 R0 [angstrom] 0.9419 &END BOND &BEND ATOMS Hw Ow Hw KIND HARMONIC K 0.139998 THETA0 [deg] 107.40 &END BEND @ENDIF</div></div>