<div dir="ltr">Hi,<div><br></div><div>Thank you for all the help/suggestions. So I decided to simply run the simulation with a fully solvated box. I also included smearing, as this is a metallic system. To set the timestep equal to 1fs I used the deuterium mass for the hydrogen atoms (I also tried with the normal H-atom and 0.5 fs timestep). The calculations always start with no problem but also always stop at a certain MD step. The error file says it is a segmentation fault, but I have no clue what this implies: is it my input file, or some compilation error (I never compiled cp2k on my own). I included my geometry file - <a href="http://frame.xyz/" target="_blank">frame.xyz</a> the <a href="http://cp2k.in/out" target="_blank">cp2k.in/out</a> and the error file. Would appreciate any suggestions.</div><div><br></div><div>Greetings,</div><div>Anna</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Am Sa., 20. Apr. 2019 um 11:29 Uhr schrieb Anna Anic <<a href="mailto:aan...@gmail.com">aan...@gmail.com</a>>:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Hi,<div><br></div><div>Thank you for all the help/suggestions. So I decided to simply run the simulation with a fully solvated box. I also included smearing, as this is a metallic system. To set the timestep equal to 1fs I used the deuterium mass for the hydrogen atoms (I also tried with the normal H-atom and 0.5 fs timestep). The calculations always start with no problem but also always stop at a certain MD step. The error file says it is a segmentation fault, but I have no clue what this implies: is it my input file, or some compilation error (I never compiled cp2k on my own). I included my geometry file - <a href="http://frame.xyz" target="_blank">frame.xyz</a> the <a href="http://cp2k.in/out" target="_blank">cp2k.in/out</a>, the <a href="http://trajectrory.xyz" target="_blank">trajectrory.xyz</a>, <a href="http://force.xyz" target="_blank">force.xyz</a> and the error file. Would appreciate any suggestions.</div><div><br></div><div>Greetings,</div><div>Anna</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Am Fr., 22. März 2019 um 19:28 Uhr schrieb Travis <<a href="mailto:polla...@gmail.com" target="_blank">polla...@gmail.com</a>>:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>Hi,</div><div><br></div><div>The potentials you apply are only zero at their nodes (z=8 and z=55) and they are non-zero everywhere else in your box, explaining the windswept appearance of your water hydrogens. You would need to supply a cube file with these potentials evaluated numerically at each grid point for z < 8 and z > 55 and zero in between. Even so, the bilayer will decompose before making contact with the barrier, so I'm not sure this approach is better than just simulating with the aqueous phase as you did previously. You can also simulate with 2D PBC using the wavelet solver and place a 2 nm slab of water on only one side of the oxide (then a bit of vacuum for the densities to trail to ~0 near the boundary in the non-periodic axis), see page 66 of <a href="https://tel.archives-ouvertes.fr/tel-01839221/document" target="_blank">https://tel.archives-ouvertes.fr/tel-01839221/document</a> for an example of what this setup looks like.</div><div><br></div><div>-T</div><div><br></div><div><br><br>On Thursday, March 21, 2019 at 3:52:02 PM UTC-4, Anna Anic wrote:</div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">PS: I appended the corresponding output file and <a href="http://force.xyz" rel="nofollow" target="_blank">force.xyz</a><div>This was just a 30 min. test run to check if the parameters work<br><div><br></div></div></div><br><div class="gmail_quote"><div dir="ltr">Am Do., 21. März 2019 um 20:43 Uhr schrieb Anna Anic <<a rel="nofollow">a...@gmail.com</a>>:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Dear Marcella,<div><br></div><div>Thank you very much for the quick response. </div><div>To answer your questions:</div><div>1. Before I did the MD part I performed energy convergence test for the slab thickness and size, then a geometry optimization for the bulk and slab with and without adsorbate with a 20 Angstrom thick vacuum level</div><div><br></div><div>But I just realized that the coordinates in the input file are not from the optimized structure. But I think that the water bilayer still should not behave like this</div><div><br></div><div>2. All structures converged during these calculations and surface free energies (SFE) are similar to reference values from other works</div><div>3. I did not calculate the binding energy for water molecules. Is this relevant if the SFE values seem to be oke?</div><div>4. The layers of the water molecules should represent the water structure in the Double Layer of an electrode/electrolyte interface. I took the geometry of the bilayer from a paper, that claims that the water at the interface is ice-like. I want to check if it stays that way (probably not) and sample as many water orientations (as this changes the work function) as possible.</div><div>5. Concerning the MD part I did not anneal the structure slowly. But I did the same calculation with a fully solvated system, so the whole vacuum region was filled with water, and the movement of the water molecules seemed reasonable. </div><div>What temperature steps for the anneling would you suggest ?</div><div>And do you maybe have some example input block for that?</div><div><br></div><div>In general, do you think that the external potential with the parabolic function is oke to keep the water molecules within a certain region from the surface?</div><div><br></div><div><br></div><div><br></div><div>Greetings,</div><div>Anna</div><div><br></div></div><br><div class="gmail_quote"><div dir="ltr">Am Do., 21. März 2019 um 17:00 Uhr schrieb Marcella Iannuzzi <<a rel="nofollow">ma...@gmail.com</a>>:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><br></div>Dear Anna, <div><br></div><div><br></div><div>which type of calculations did you carry out in preparation of the MD run?</div><div>Is your DFT model reproducing the correct lattice structure?<br></div><div>Did you optimise the slab and check that the electronic structure of the bulk and at the surface is what expected from reference data?</div><div>I do not know this specific system, but the parameters in the DFT section seem to be not very well tuned. Is the SCF converging? Do you get reasonable forces?<br></div><div><br></div><div>Did you model the adsorption of individual water molecules to evaluate the binding energy and compare to references? </div><div>Should the layer of molecules be ice-like or do you want to simulate liquid water?</div><div>If the coordinates are obtained from some structural data or geometry optimisation, did you anneal slowly from 0K to the final temperature?</div><div>Why are you running NPE? Did you already thermalised at constant volume?</div><div><br></div><div>Kind regards</div><div>Marcella</div><div><br></div><div><br></div><div><br><br>On Thursday, March 21, 2019 at 3:55:19 PM UTC+1, Anna Anic wrote:<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Hey everybody,<div><br></div><div>I am trying to do Ab-initio MD for a metal-water interface. I am using a symmetric slab and positioned a bilayer of water at both surface sites. </div><div>Now, if I run the simulation my water layer just decomposes into the vacuum region. That is why I would like to implement an vacuum potential barrier at some point in the z-direction to prevent my water from just "flying away".</div><div>I tried using the &EXTERNAL_POTENTIAL keyword and defined a harmonic function as a barrier.</div><div>But when I run the simulation the hydrogen atoms from the H2O-bilayer are "blown away".</div><div>The question is, what did I do wrong? Is the idea legit? Does anyone have a proposal how to fix the problem or how to do it differentey.</div><div><br></div><div>Greetings,</div><div><br></div><div>Anna</div><div><br></div><div>(I attached my input file,geometry and trajectory )</div></div></blockquote></div></div> <p></p> -- <br> You received this message because you are subscribed to the Google Groups "cp2k" group.<br> To unsubscribe from this group and stop receiving emails from it, send an email to <a rel="nofollow">cp...@googlegroups.com</a>.<br> To post to this group, send email to <a rel="nofollow">c...@googlegroups.com</a>.<br> Visit this group at <a href="https://groups.google.com/group/cp2k" rel="nofollow" target="_blank">https://groups.google.com/group/cp2k</a>.<br> For more options, visit <a href="https://groups.google.com/d/optout" rel="nofollow" target="_blank">https://groups.google.com/d/optout</a>.<br> </blockquote></div> </blockquote></div> </blockquote></div> <p></p> -- <br> You received this message because you are subscribed to the Google Groups "cp2k" group.<br> To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:cp...@googlegroups.com" target="_blank">cp...@googlegroups.com</a>.<br> To post to this group, send email to <a href="mailto:cp...@googlegroups.com" target="_blank">cp...@googlegroups.com</a>.<br> Visit this group at <a href="https://groups.google.com/group/cp2k" target="_blank">https://groups.google.com/group/cp2k</a>.<br> For more options, visit <a href="https://groups.google.com/d/optout" target="_blank">https://groups.google.com/d/optout</a>.<br> </blockquote></div> </blockquote></div>