From: Fco. J. M. <mo...@un...> - 2004-11-29 15:49:29
|
Just my two (euro)cents about this issue. I have recently moved from JMolAppletPre10 to pre 20. An .xyz animation which previously was fine now show some bonds broken. The animation should show Berry's pseudorotation using a sequence of coordinates of AsF5 in .xyz format as a model of fluxional behaviour. For example the first step is: -------- 6 Frame 1 F 0.0000 0.0000 -0.9570 As 0.0000 0.0000 0.7034 F -1.4343 -0.0001 1.5387 F 1.4343 0.0001 1.5387 F -0.0001 1.6817 0.7002 F 0.0001 -1.6817 0.7004 ---------- The problem is that only three As-F bond are shown instead of five, just the three F which are closer to As. As the animation advances the bonding changes but it keeps showing just three bonds. I guess that some internal parameters (either the minimum bond distance, the covalent radii or the coordination number...) in JMolApplet have been changed in recent releases. I have been reading this post about "atom types" and I have found also that there was an "autobonding of oxygen" thread in the developers' list which discuss around a similar question. I have read in some post about a pathological case of hundred of bonds around an atom. Obviously there is some chemical problem which such a model, I cannot think about any sensible structure of this kind. Only in strongly dissordered X-Ray structural data this kind of things could have any meaning (I think). My personal view is that everything should be kept as simple as posible and the autobonding feature should be based just in the sum of covalent radii. For inorganic and organometallic chemistry structures it is not very convenient to limit the number of coordination based on the type of atom. For example, more than 4-coordinate carbon atom is no such rare beast in inorganic chemistry either in solid state carbides, in organometalic compounds carbide ligands or in carbon monoxide acting as a bridge between metal atoms, not to speak about bridging halides, bridging hydrides and so on. F.J. Modrego -- Dr. F.J. Modrego Department of Inorganic Chemistry Facultad de Ciencias University of Zaragoza 50009 ZARAGOZA SPAIN Tel <34>-976-762288 Fax <34>-976-761187 E-mail: mo...@un... |
From: Miguel <mi...@jm...> - 2004-11-29 16:50:42
|
Francisco en Espana escribio: > Just my two (euro)cents about this issue. > I have recently moved from JMolAppletPre10 to pre 20. An .xyz > animation which previously was fine now show some bonds broken. =5Bsnip=5D > The problem is that only three As-F bond are shown instead of five, > just the three F which are closer to As. =5Bsnip=5D > I guess that some internal parameters (either the minimum > bond distance, the covalent radii or the coordination > number...) in JMolApplet have been changed in recent releases. I recently *enhanced* the bonding table by introducing maximum limits on the number of bonds in the autobonding algorithm. I took the numbers from= openbabel. Indeed, 'As' has a limit of 3. This *enhancement* has caused a number of difficulties. It will be remove= d in the next release. > I have read in some post about a pathological case of > hundred of bonds around an atom. To minimize the impact caused by this type of pathological problem the modified code will have an upper limit of 20 bonds per atom in the 'autobonding' code. =21Adios=21 Miguel |