This experiment is a computational supplement to Part B of the tin chemistry described in "Synthesis and Technique in Inorganic Chemistry" (Exp 7; see below for the complete citation).* Students will optimize and compute IR spectra for the cis and trans and corresponding linkage isomers of tetrachlorbis(dimethylsulfoxide) tin(IV). A comparison of experimental (IR spectra) and computational data (enthalpies of formation; IR spectra) will aid them in determining the most likely product of this simple synthesis and in identifying the S-O vibrations in their experimental spectrum.
*G. S. Giorlami, T. B. Rauchfuss, R. J. Angelici “Synthesis and Technique in Inorganic Chemistry: A Laboratory Manual”, Third Edition
Computational software package.
Computational time is minimized in this experiment by using a semi-empirical method (PM3).Unfortunately this also means that the IR spectra are more qualitative than quantitative.However, given that four structures must be studied, this was deemed to be a reasonable compromise. Since this is a computational exercise, computer access during typical laboratory sessions is required.Since I run my lab in the round-robin style, this was not a significant issue (1-3 computers were needed in any given week).
Students had some difficulty drawing some of the isomers.The ease of drawing the isomers depends in part on the GUI you have available but is also complicated by trying to place two DMSO molecules cis to each other.This is complicated further by students having trouble visualizing the structure of DMSO--apparently because they are rather unfamiliar with sulfur occuring in a molecule.The computer based visualization helps immensely with the student visualization, forcing students to do more than treat the synthesis as a cookbook experiment (with little thought as to the details). Students should also be cautioned to check local atom geometries.There were a number of instances where students managed five-coordinate carbon (causing the optimization to fail in most cases) or they forgot about/ignored the sulfur lone pairs (this is where you may see distintive differences between the different computational packages since most seem to be designed with organic chemistry in mind).In addition one or two of the cis isomers failed to geometry optimize even when the structures were deemed to be correct.Slight changes in the geometry may result in a reasonable optimization, but I was more interested in students coming up with a reasonable explanation as to why the cis structures were unlikley products.So when a structure failed to optimize I coached them to look carefully at their structure and identify where problems might arise.I told them I was perfectly happy with a failed optimization if I had very good reason to believe the optimzation would be diffucult. The semieprical method (PM3) chosen was decided upon merely for speed of computation.Ab-initio methods, while likely giving better results, would take far too much time and in the end don't seem to be necessary.The trans geometries end up with the lowest energy of formation, but the difference between the S vs O linkage isomer seems to be well within computational error.Thus while the O-linkage is the lowest energy structure, students must really compare their experimental IR spectra with the computational spectra to confirm this assignment. I have also found this exercise to be a good way for students to visualize specific vibrations, helping them identify the S-O in their experimental spectrum. A good GUI is required to input the structures into computational software relatively painlessly. I do not recommend using the WebMO interface for drawing the structures, but once a structure has been drawn it be imported with only minor difficulties.