Rep:Mod:AW611
Experiment 1-C: Part 1
Conformational analysis using molecular mechanics
Dimerisation of cyclopentadiene and the hydrogenation of the dimers
Theoretically, when cyclopentadiene dimerises, by a Diels-Alder reaction, it can form either the exo 1 or the endo 2 dimer as shown in Fig. 1. In reality, upon dimerisation, only the endo dimer is formed. The aim of this part of the experiment was to use molecular mechanics in order to predict the relative energies of the exo and endo dimers, and using this information it was possible to identify whether the dimerisation was under kinetic or thermodynamic control.
The endo dimer can undergo hydrogenation to form one of two possible dihydro derivatives where either the cyclopentene is hydrogenated 3 or the norbornene is 4 as shown in Fig. 1. Using similar molecular mechanics calculations as above, a thermodynamic prediction regarding the stability of the two dihydro derivatives was made, and this identified which of them was more thermodynamically stable, and therefore more likely to form. Again this information made it possible to find out whether the hydrogenation was under kinetic or thermodynamic control.
| Molecule 1 | Molecule 2 | Molecule 3 | Molecule 4 | |
|---|---|---|---|---|
| Stretch (kcal/mol) | 1.2850 | 1.2508 | 1.2348 | 1.0965 |
| Bend (kcal/mol) | 20.5805 | 20.8478 | 18.9384 | 14.5243 |
| Stretch-Bend (kcal/mol) | -0.8380 | -0.8358 | -0.7609 | -0.5494 |
| Torsion (kcal/mol) | 7.655 | 9.5109 | 12.1241 | 12.4974 |
| Non-1,4 VDW (kcal/mol) | -1.4173 | -1.5438 | -1.5018 | -1.0700 |
| 1,4 VDW (kcal/mol) | 4.2333 | 4.3199 | 5.7290 | 4.5126 |
| Dipole/ Dipole (kcal/mol) | 0.3775 | 0.4476 | 0.1631 | 0.1406 |
| Total E (kcal/mol) | 31.8764 | 33.9975 | 35.9266 | 31.1520 |
ChemBio3D was used to draw out the exo and endo dimers, and then their energies were minimised using the MM2 force field. This then provided a breakdown of the energy contributions for each molecule, and their total energies. By comparing the energies of 1 and 2 it is clear that the only significant difference was in the torsion. The endo dimer has greater torsional strain due to the close proximity of the two rings. This greater torsion leads to the endo dimer having a greater total energy, thus making the exo dimer more thermodynamically stable. However, the product formed is the endo dimer [1] this means that the dimerisation is under kinetic control. The endo dimer forms more quickly because the molecules line up, overlapping p orbitals from the two rings interact, lowering the energy of the transition state. The exo dimer is formed by an end to end approach where the orbitals are too far apart for this stabilising interaction to take place. The findings of these calculations agree with those in literature where, through similar calculations, the endo dimer was found to be more thermodynamically stable. [2]
The two dihydro derivatives were also drawn out in ChemBio3D and minimised using the MM2 force field. From the energy contributions calculated, the only parameter with significant differences between 3 and 4 is bending, this was ~4 kcal/mol. The bond angles in 4 are a more favourable, as there is less strain in the system, leading to the molecule having the lower total energy and therefore making it the most thermodynamically stable product. The bond angles around the sp2 hybridised carbon are closer to the optimal 120o in 4 compared to 3 (127.1o in 3 and 124.4o in 4). Product 4 is reported as the major product in literature [3] therefore the hydrogenation is under thermodynamic control, as the lower energy product is formed predominantly.
The synthesis of Taxol, used in the treatment of ovarian cancer, proceeds via an intermediate, which is one of the two atropisomers 9 and 10 shown in Fig. 2. The aim of this part of the experiment was to use molecular modelling to determine which of the atropisomers was most stable, and therefore likely to be the identity of the intermediate. Another aim was to use the information from these calculations in order to explain why the alkene group present in the molecule reacts slowly.
The cyclohexane motif in the molecule leads to there being 4 possible conformations for each of molecules 9 and 10, 2 chairs and 2 twist-boats. Since twist-boats are always higher in energy they can be discounted from the search for the lowest energy conformer. Both chair conformers were found for 9 and 10 and their energies calculated , the conformer with the lower energy was used in each case to identify whether 9 or 10 was more thermodynamically stable, this is shown in Table 2.
Again, ChemBio3D was used to draw out the two atropisomers and their conformers, and using the MM2 force field, their energies were minimised. The total energy and the individual contributions for the lowest energy conformers of 9 and 10 were calculated, and these values are shown in Table 3. The parent alkanes were also drawn out for the lowest energy conformers of 9 and 10 and their energies minimised and calculated in order to investigate why the alkene in 9 and 10 is so unreactive.
| Molecule 9 (A) | Molecule 9 (B) | Molecule 10 (A) | Molecule 10 (B) | |
|---|---|---|---|---|
| Total E (kcal/mol) | 47.8395 | 58.3858 | 42.6828 | 52.5428 |
For both molecules 9 and 10 there is one chair conformer that is significantly lower than the other, these were then identified as the lowest energy conformers and were used to compare against one another to find which was more stable.
| Molecule 9 | Molecule 10 | |
|---|---|---|
| Stretch (kcal/mol) | 2.7846 | 2.6205 |
| Bend (kcal/mol) | 16.5412 | 11.3390 |
| Stretch-Bend (kcal/mol) | 0.4304 | 0.3432 |
| Torsion (kcal/mol) | 18.2512 | 19.6720 |
| Non-1,4 VDW (kcal/mol) | -1.5525 | -2.1618 |
| 1,4-VDW (kcal/mol) | 13.1094 | 12.8721 |
| Dipole/ Dipole (kcal/mol) | -1.7248 | -2.0023 |
| Total E (kcal/mol) | 47.8395 | 42.6828 |
From the data in Table 3, the values for most parameters are very similar for 9 and 10, the only energy contribution with a significant difference is bending with 10 being ~5 kcal/mol lower than 9. This means that the bonding angles are more favourable in 10, the bond angles around the sp2 hybridised carbonyl C (i.e. O-C-C) are ~120o in 10, but the angles are ~115o in 9 and this is unfavourable. Therefore there is less steric clashing in 10 and hence it is clear that it is the more stable atropisomer.
Hyperstable alkenes are olefins that have less strain than their parent alkane[4], and they often contain a bridgehead. The hyperstability of the alkenes arises from the increase in unfavorable interactions[5] of the vicinal and transannular hydrogens, which leads to the alkenes being more energetically stable than the alkane, thus making the C=C bond particularly unreactive. The hydrogenated parent alkanes of molecules 9 and 10 were modelled: Hydrogenated Molecule 9 and Hydrogenated Molecule 10 and the bond angles at the base of the bridgehead (where the alkene was located prior to hydrogenation) were 118.7o and 122.7o respectively, since this carbon is sp3 hybridised, both have very unfavourable bond angles, causing strain in the molecule. From the Jmol's it is also possible to see steric clashing of the bridgehead methyls with hydrogens on the ring, these are also unfavourable, and further destabilise the parent alkanes relative to 9 and 10, confirming that they are hyperstable olefins.
Spectroscopic simulations of Taxol intermediates using quantum mechanics
Molecules 17 and 18 (shown in Fig. 3) are also Taxol intermediates, closely related to molecules 9 and 10 mentioned above. Using quantum mechanics it is possible to calculate the 1H and 13C NMR spectra for the molecule, in this case 18 was selected, as it is a derivative of the more stable isomer, molecule 10. The calculated spectra were compared to literature values[6] to see if they had been assigned correctly in the original papers.
Again, there are many conformers possible for 18, so another element to this investigation was to find out whether different conformers produced significantly different spectra. The conformers chosen in this case were one that contained a chair and one that contained a boat. Their energies were minimised using the MMFF94s force field in Avagadro then the spectra were calculated[7][8] using Gaussian. The chemical shifts from the spectra shown in Tables 4 and 5.
NB: The literature 1H NMR: (300 MHz, C6D6) and 13C NMR: (75 MHz, C6D6)
| Lierature Data | QM Simulation for Chair | QM Simulation for Boat |
|---|---|---|
| 5.21 ppm (m, 1H) | 6.00 ppm (1H) | 5.27 ppm (1H) |
| 3.00-2.70 ppm (m, 6H) | 3.16-3.12 ppm (2H) | 3.29 ppm (1H) |
| 2.70-2.35 ppm (m, 4H) | 3.00 ppm (1H) | 3.19 ppm (1H) |
| 2.20-1.70 ppm (m, 6H) | 2.94-2.92 ppm (2H) | 3.06 ppm (2H) |
| 1.58 ppm (t, J=5.4 Hz, 1H) | 2.81 ppm (2H) | 2.86-2.84 ppm (2H) |
| 1.50-1.20 ppm (m, 3H) | 2.62 ppm (1H) | 2.52 ppm (1H) |
| 1.10 ppm (s, 3H) | 2.56 ppm (1H) | 2.45 ppm (1H) |
| 1.07 ppm (s, 3H) | 2.48 ppm (1H) | 2.35 ppm (1H) |
| 1.03 ppm (s, 3H) | 2.35-2.33 ppm (2H) | 2.28-2.27 ppm (2H) |
| - | 2.28 ppm (1H) | 2.14 ppm (1H) |
| - | 2.01-2.00 ppm (2H) | 1.97-1.92 ppm (2H) |
| - | 1.88-1.83 ppm (2H) | 1.83-1.82 ppm (2H) |
| - | 1.66 ppm (1H) | 1.69-1.66 ppm (2H) |
| - | 1.59 ppm (1H) | 1.58 ppm (1H) |
| - | 1.52 ppm (2H) | 1.53 ppm (1H) |
| - | 1.37 ppm (1H) | 1.47 ppm (1H) |
| - | 1.31 ppm (1H) | 1.40 ppm (1H) |
| - | 1.25-1.21 ppm (2H) | 1.34 ppm (1H) |
| - | 0.98-0.92 ppm (3H) | 1.16 ppm (1H) |
| - | 0.63 ppm (1H) | 1.06-1.02 ppm (3H) |
| - | - | 0.97-0.94 ppm (2H) |
From the 1H NMR data shown in Table 4. it is clear that the boat conformer has a spectrum that matches very closely. The alkene signal (5.27 ppm) is very close to the reported literature value, and almost all of the signal lie with 0.20ppm of the literature values. Also the degeneracy of the signals matches the signals in literature quite closely. The multiplets mentioned in the literature are probably due to a number of separate signals that lie close to one another on the spectrum and have overlapped. After the distinctive singlet at ~5.2ppm, the signals that correspond to the next 6 hydrogens (3.29-2.84ppm) lie in the same region as the 6H multiplet in the literature. The signals that correspond to the following 10 hydrogens (2.52-1.82ppm) lie closely to the 4H and 6H multiplets in the literature. The signals for the next 4 hydrogens (1.69-1.53ppm)correspond to the triplet 1H and singlet 3H. With the remaining signals (corresponding to the 9 most shielded protons) correspond to the final 3 singlets 9H in the literature.
The 1H data for the chair conformer agrees much less closely than that of the boat. The singlet due to the alkene H is at 6.00ppm which is much higher than predicted by the literature. Also, there are a few other signals e.g. the most shielded proton at 0.63ppm which do not match the literature particularly well. The degeneracy of the signals only fit that seen of the multiplets in literature very loosely, and signals that are expected to overlap are not always very close. The poor fit suggests that either the literature 1H NMR was measured for the boat, or the chair conformer calculated in this invesitgation was not the lowest energy chair.
Also, by comparing the 1H NMR spectrum of the chair conformer with the boat, it can be seen that the chemical shifts of some signals do differ by a large amount (~0.7ppm for the alkene proton) but most signals remain quite similar. The pattern of signals is also very close, but there are some subtle differences. Since the different conformations lead to the hydrogens lying in different positions it is logical that the spectra differ slightly.
| Literature Data | QM Simulation for Chair | QM Simulation for Boat |
|---|---|---|
| 211.49 ppm | 212.94 ppm | 210.58 ppm |
| 148.72 ppm | 147.91 ppm | 148.35 ppm |
| 120.90 ppm | 120.12 ppm | 118.90 ppm |
| 74.61 ppm | 93.17 ppm | 88.71 ppm |
| 60.53 ppm | 65.83 ppm | 67.85 ppm |
| 51.30 ppm | 54.94 ppm | 55.87 ppm |
| 50.94 ppm | 54.93 ppm | 55.45 ppm |
| 45.33 ppm | 49.57 ppm | 49.94 ppm |
| 43.28 ppm | 48.00 ppm | 48.00 ppm |
| 40.82 ppm | 45.70 ppm | 44.04 ppm |
| 38.73 ppm | 44.05 ppm | 41.97 ppm |
| 36.78 ppm | 41.24 ppm | 34.35 ppm |
| 35.47 ppm | 38.63 ppm | 35.19 ppm |
| 30.84 ppm | 33.62 ppm | 30.92 ppm |
| 30.00 ppm | 32.48 ppm | 28.87 ppm |
| 25.56 ppm | 28.31 ppm | 28.59 ppm |
| 25.35 ppm | 26.41 ppm | 25.73 ppm |
| 22.21 ppm | 24.47 ppm | 25.16 ppm |
| 21.39 ppm | 24.07 ppm | 23.39 ppm |
| 19.83 ppm | 22.48 ppm | 20.98 ppm |
From Table 5. it is clear that the 13C data for both the boat and chair conformers are very similar, both to one another, and to the literature values. Almost all signals are within 5ppm of the literature data. There is, however, the signals that relate to the carbons bonded to sulfur atoms, their chemical shifts are slightly further away from the literature values (these are the carbons with chemical shifts 44.05ppm and 41.24ppm for the chair). Another is the signal that corresponds to the carbon bonded directly to two sulfurs (93.17 ppm in chair and 88.71ppm in boat conformer) is very far from the literature values. This may be due to spin-orbital coupling error, caused by the heavy sulfur atoms, and could be corrected.
The 13C spectra for the boat and chair are very similar in this case, the chemical shifts are very close together, generally within ~2ppm. Therefore the carbons must remain in very similar environments despite the change in conformation, since the carbons still have the same connectivity in both conformers.
References
- ↑ Cristol, S. J.; Seifert, W. K.; Soloway, S. B. J. Am. Chem. Soc. 1960, 82, 2351
- ↑ W. L. Jorgensen, D. Lim, J. F. Blake; J. Am. Chem. Soc., 1993, 115 (7), 2936-2942. DOI:10.1021/ja00060a048
- ↑ G. Liu , Z. Mi , L. Wang , and X. Zhang, '"Kinetics of Dicyclopentadiene Hydrogenation over Pd/Al2O3 by catalyst", Ind. Eng. Chem. Res., 2005, 44(11), 3846-3851. DOI:10.1021/ie0487437
- ↑ W. F. Maier, P. Von Rague Schleyer, J. Am. Chem. Soc., 1981, 103, 1891. DOI:10.1021/ja00398a003
- ↑ S. Vasquez, P. Camps; "Chemistry of pyramidalized alkenes" Tetrahedron, 2005, 61, 5147-5208. DOI:10.1016/j.tet.2005.03.055
- ↑ L. A. Paquette, N. A. Pegg, D. Toops, G. D. Maynard and R. D. Rogers, J. Am. Chem. Soc., 1990, 112, 277-283. DOI:10.1021/ja00157a043
- ↑ A. Wadsworth,"Taxol intermediate 18 chair NMR" D-Space, 2013 DOI:10042/25762
- ↑ A. Wadsworth, "Taxol intermediate 18 boat NMR" D-Space, 2013 DOI:10042/25763
Experiment 1-C: Part 2
Crystal structures of Shi and Jacobsen pre-catalysts
Crystal structures for the Shi and Jacobsen pre-catalysts, 21 and 23 respectively, were found by searching the Cambridge Crystal Database using Conquest. This was done in order to measure the C-O bond lengths of the anomeric centres in the Shi pre-catalyst, and examine the close approach of t-butyl groups in the Jacobsen pre-catalyst. The selectivity of each of the catalysts can also be explained by investigating the crystal structures.
Shiprecusor21 |
Firstly, by looking at the crystal structure it is possible to explain why the Shi catalyst only catalyses the reaction of (E) alkenes. The dioxirane sits close to the plane of the cyclohexane ring (where the carbonyl sits in pre-catalyst 21) and so when the alkene attacks, the groups substituted onto it will experience steric hindrance from one of the five membered rings. One of these rings points up, and the other down, relative to the cyclohexane part, and therefore only a trans alkene would be able to reach the dioxirane oxygens without experiencing huge amounts of steric hindrance, cis alkenes could not. For the anomeric centre located on the five member ring, there is one O-C bond that is 0.141nm and the other is 0.144nm, this is due to the anomeric effect, hyperconjugation of the oxygen LP with a C-O σ* orbital stabilises the molecule, this weakens one of the bonds, thereby increasing the bond length. At the other anomeric centre, located on the six membered ring, the bond lengths are the same, this is because the hetero atom is forced into equatorial as other substituents on the ring favour axial positions far more than itself. This means there is no hyperconjugation of the lone pair of oxygen, due to the lack of app arrangement of orbitals when the heteroatom is equatorial, thus the bond lengths are the same.
Jacobsen23 |
The hydrogens of the t-Butyl groups (highlighted in yellow) are ~0.260nm apart, this is in the region for attractive Van der waals interactions. The stabilised close approach of the t-Butyl groups effectively blocks attack of the alkene from that position. Also side approaches are blocked by the other t-Butyl groups, hence the incoming alkene must approach by passing over the cyclohexane part of the molecule. This leads to the selectivity of the Jacobsen catalyst, and also limits the incoming alkenes to being the Z isomer to avoid steric hindrance.
Calculated NMR of epoxides
The alkenes chosen to undergo epoxidation were styrene and trans-β-methylstyrene see Fig. 1. The epoxides of these alkenes ((R)-styrene oxide, (S)-styrene oxide, (R,R)-trans-β-methylstyrene oxide and (S,S)-trans-β-methylstyrene oxide) were drawn out in Avagadro and their energies minimised using the MMFF94s force field. The 1H and 13C NMR spectra for both epoxides were calculated[1][2][3][4] as part of the information to aid in assigning the epoxides, and the resulting data is displayed below. This data was compared with literature spectra ( 1H NMR for styrene oxide,[5] 13C NMR for styrene oxide,[6] 1H and 13C NMR for trans-β-methylstyrene oxide[7])
| Computed 1H | Literature 1H | Computed 13C | Literature 13C |
|---|---|---|---|
| 7.51-7.45ppm (4H) | 7.4-7.3ppm (5H) | 135.13ppm | 137.6ppm |
| 7.30ppm (1H) | 3.83ppm (1H) | 124.14ppm | 128.4ppm |
| 3.66ppm (1H) | 3.12ppm (1H) | 123.42ppm | 128.2ppm |
| 3.12ppm (1H) | 2.77ppm (1H) | 122.96ppm | 125.5ppm |
| 2.54ppm (1H) | - | 118.27ppm | 52.4ppm |
| - | - | 54.06ppm | 51.2ppm |
| - | - | 53.48ppm | - |
The spectra for the two enantiomers were identical so only one set of data has been shown. From Table 1, it is clear that both the 1H and 13C calculated NMR spectra match the literature values for styrene oxide. The computed chemical shifts are within ~0.2ppm for 1H and ~4ppm for 13C, and the degeneracy is the same for both. In the literature 13C there is one less signal, suggesting one less chemically distinct carbon environment, and this is due to two carbons in the same environment, which is seen as two signals very close to one another. The close match of the literature data to the computed spectra confirms that the correct epoxide has been modeled, the calculations also displayed that the different enantiomers do not produce differing spectra.
| Computed 1H | Literature 1H | Computed 13C | Literature 13C |
|---|---|---|---|
| 7.50-7.48ppm (3H) | 7.27ppm (5H) | 134.98ppm | 137.7ppm |
| 7.42ppm (1H) | 3.55ppm (1H) | 124.08ppm | 128.3ppm |
| 7.31ppm (1H) | 3.12ppm (1H) | 123.33ppm | 127.9ppm |
| 3.41ppm (1H) | 1.43ppm (3H) | 122.80ppm | 125.4ppm |
| 2.79ppm (1H) | - | 122.73ppm | 59.4ppm |
| 1.68ppm (1H) | - | 118.49ppm | 58.9ppm |
| 1.59ppm (1H) | - | 62.31ppm | 17.8ppm |
| 0.72ppm (1H) | - | 60.58ppm | - |
| - | - | 18.84ppm | - |
Again the two enantiomers produced identical spectra so only one set of data has been shown. Table 2 shows that again the computed 13C spectrum very closely matches the literature data, the chemical shifts for 13C are ~4ppm and again the differeing number of signals is due to carbons in the same chemical environment. In this case the 1H spectrum matches literature less closely, the signals still form the same patterns i.e. signals corresponding to 5H in ~7ppm region, then two 1H signals in the ~3ppm region, and three 1H signals in the ~1ppm region. Although the chemical shifts match the literature less closely than for styrene oxide.
Since both the 1H and 13C NMR spectra are the same for both enantiomers (in both styrene oxide and trans-β-methylstyrene) it is not possible to distinguish between the enantiomers, and therefore it's impossible to assign the configuaration of the epoxides using the NMR data.
Assigning absolute configuration of epoxides
Since two enantiomers can be formed for each epoxide, styrene oxide (using the Shi catalyst) and trans-β-methylstyrene (using the Jacobsen catalyst), it is important to assign which configuration of the epoxides is actually formed. Using a variety of physical data it is possible to assign the configuration with a large degree of certainty.
Computed optical rotations
The output files (optimised) from the NMR calculations were then reused to calculate the ORP[8][9][10][11] of the epoxides using light of incident wavelenth 589nm. These jobs were run on the HPC.
| (R)-styrene oxide | (S)-styrene oxide | (R,R)-trans-β-methylstyrene oxide | (S,S)-trans-β-methylstyrene oxide |
|---|---|---|---|
| -30.35 deg | +30.55 deg | +47.08 deg | -46.96 deg |
Searches for literature values of ORP were inconclusive, both positive and negative ORP values had been associated with each enantiomer, making the data unreliable. This meant it was not possible to assign the configuration of the epoxides obtained using the optical rotation data.
Comparison of transition states
The transition states were modelled to find the Gibbs free energies of the transition states for both enantiomers of trans-β-methylstyrene oxide (with Shi catalyst) and of phenylprop-1-ene (with Jacobsen catalyst). By comparing the Gibbs free energies of each enantiomer, for a given transition state, it was possible to determine the equilibrium constant for the conversion of the enantiomers, and thus work out which was in enantiomeric excess. Since there was more than one transition state for both trans-β-methylstyrene and phenylprop-1-ene epoxidation the lowest energy transition state was used in each case, as this is the most likely pathway for the reaction to proceed. The equilibrium constant K was determined using ΔG= -RT ln(K).
| Change in ΔG for trans-β-methylstyrene (R,R to S,S interconversion) | Equilibrium constant (K) for trans-β-methylstyrene (R,R to S,S interconversion | Change in ΔG for phenylprop-1-ene (S,R to R,S interconversion) | Equilibrium constant (K) phenylprop-1-ene (S,R to R,S interconversion) |
|---|---|---|---|
| -20.218977 kJ/mol | 3486.652116 | -22.3141262 kJ/mol | 8118.640311 |
The chosen transition state is the lowest energy transition state, this corresponds to the state where the oxygen transfer occurs at the dioxirane oxygen, which is located in the equatorial position on the cyclohexane ring. The equilibrium constant for the interconversion of R,R to S,S trans-β-methylstyrene is large (3486.652116) and this shows that the S,S enantiomer is dramatically favoured. The enantiomeric excess for this was calculated to be 99.97% in favour of S,S which confirms this. The fact that the ΔG for this interconversion is negative also shows this, since a negative change in Gibbs free energy corresponds to a spontaneous process.Therefore it is highly likely that the enantiomer formed by epoxidation of trans-β-methylstyrene using the Shi catalyst is the S,S enantiomer.
Although phenylprop-1-ene was not one of the alkenes chosen in this experiment, the data of it's transition state using the Jacobsen catalyst can be used in order to learn which enantiomer of styrene oxide is preferred. The equilibrium constant for the conversion of the S,R to R,S epoxide was very large (8118.640311), which suggests that the R,S epoxide is favoured. The enantiomeric excess calculated for this was 99.99% in favour of R,S. Again the negative change in Gibbs free energy shows that the forward reaction is spontaneous.
Styrene oxide only has one enantiomeric centre, rather than two, but since it undergoes a very similar reaction, using the same catalyst (Jacobsen), it is possible to use the thermodynamic data above in order to assign which enantiomer of styrene oxide is formed. The R enantiomer is favoured as this corresponds to the R,S of phenylprop-1-ene epoxide's transition state, which, as shown above, has a large enantiomeric excess (99.99%).
Investigating the non-covalent interactions in the reaction transition state (NCI)
By calculating the non covalent interactions in the transition state (2nd (R,R) transition state) it is possible to visualise the intermolecular interactions that take place, and this aids in the understanding of the mechanism of the reaction.
Orbital |
The electrostatic potential isosurface that lies between a hydrogen on the five membered ring and a terminal methyl hydrogen is green, therefore indicating that it is a weakly attractive interaction. Since this is between to hydrogens, it is likely that this is an attractive Van der Waals interaction. Another interaction is also located in the 5 membered ring, this time between the lone pairs of two oxygens. The red isosurface in the centre of the 5 membered ring is due to repulsion between the oxygen lone pairs, this causes destabilisation in the ring, however this is counteracted to an extent by the Van der Waals mentioned above, along with other O-H dipole interactions, this stops the ring from buckling.
Finally, the isosurface that forms a ring, between the dioxirane oxygen and the alkene shows the interaction taking place where the bond is forming. Both red, strongly repulsive (due to oxygen lone pair-π electron repulsions) and blue, strongly attractive interactions (due to the stability of the new species formed) are present. The interactions here are of much greater complexity so it is harder to analyse this isosurface.
Investigating the electronic topology (QTAIM)
The same transition state ((R,R) 2nd transition state) was then subjected to a QTAIM analysis. This identifies points where the first derivative of the electron density is zero, these correspond to maxima, i.e. where the electron density is greatest. this can then tell us more information about the bonds, and where the electron density lies.
In terms of covalent bonds, it is clear that for C-C and C=C bonds the maximum electron density lies exactly in the middle of the two atoms. For C-H bonds the electron density lies slightly closer to the H since it is slightly more electronegative. The C-O bonds have their electron density much closer to the oxygen, due to the electronegative nature of O. These, though, are of little interest as they only confirm what is already known about these types of bond.
The more interesting interactions are the non covalent interactions between the alkene and the dioxirane oxygen (1). The electron density lies between the two molecules as this is a transition state, but shows that a bond is starting to form between the oxygen and the alkene carbons.
New Candidates
Another possible candidate for investigation is the epoxidation of 4-methyl-stilbene using the Shi catalyst. 4-methyl-stilbene oxide has an Optical Rotary Power (ORP) of -580 deg (436nm)[12]. 4-methyl-stilbene is synthetically accessible as it is readily available and can be purchased from suppliers, also the high value for ORP means that it is fairly easy to investigate.
References
- ↑ A. Wadsworth, "(R)-Styrene oxide NMR" D-Space, 2013 DOI:10042/25764
- ↑ A. Wadsworth, "(S)-styrene oxide NMR" D-Space, 2013 DOI:10042/25768
- ↑ A. Wadsworth, "(R,R)-trans-β-methylstyrene oxide NMR" D-Space, 2013 DOI:10042/25765
- ↑ A. Wadsworth, "(R,R)-trans-β-methylstyrene oxide NMR" D-Space, 2013 DOI:10042/25770
- ↑ J. M. Vega-Perez, I. Perinan, M. Vega-Holm, C. Palo-Nieto, F. Iglesias-Guerra; Tetrahedron, 2011, 67, 7057-7065. DOI:10.1016/j.tet.2011.07.014
- ↑ K. Sarma, A. Goswami, B. C. Goswami ; Tetrahedron: Asymmetry, 2009, 20, 1295-1300. DOI:10.1016/j.tetasy.2009.05.001
- ↑ H. Hiyachi, Y. Kon, K. Takumi, N. Sasagawa, Y. Ezaki, K. Sato ; Synthesis, 2012, 44 (11), 1672-1678. DOI:10.1055/s-0031-1290948
- ↑ A. Wadsworth, "(R)-styrene oxide ORP" D-Space, 2013 DOI:10042/25766
- ↑ A. Wadsworth, "(S)styrene oxide ORP" D-Space, 2013 DOI:10042/25767
- ↑ A. Wadsworth, "(R,R)-trans-β-methylstyrene oxide ORP" D-Space, 2013 DOI:10042/25769
- ↑ A. Wadsworth, "(S,S)-trans-β-methylstyrene oxide ORP" D-Space, 2013 DOI:10042/25771
- ↑ P. M. Damsette, H. Ziffer, D. M. Jerina, ; Tetrahedron, 1976, 32 (17), 2071–2074. DOI:10.1016/0040-4020(76)85110-1