ChemWiki - User contributions [en]

Talk:Mod:ms7109 module1

2011-12-02T16:17:11Z

Mjhughes:

Q1: Your energy values are all spot on although it was not necessary to give data to so many decimal places: these are not precisely measured quantities and there is also some degree of error due to random fluctuation from one calculation to the next or one laptop to the next. The analysis of strain contributions is good. Thermodynamic vs kinetic control is correctly described for these examples; as you say for the hydrogenated compounds – you know which product would be formed under thermodynamic control, but for kinetically controlled reactions information about the transition states must be obtained. It is good that you have included some references here, but it should be noted that while Baldwin provided an explanation for endo control, the original “determination” that the endo isomer is preferred was made empirically at the turn of the century by Diels and Alder.

Q2. The approach you have taken makes sense (looking at the part of the structure for which various stable conformations are known). Most of your structures and energy values are good. Your calculation for the up-chair isomer does not have the correct structure – the geometry of the double bond is incorrect. It is important to always check the structure after minimisation to ensure it still has the correct configuration (this can easily be compromised when the structure is tweaked to find different starting points). It would have been good to hear about how else you minimised the structure – were there alternative conformations for the central medium ring? Even if this was not the case, it is worth stating that. The strain in hydrogenation is well described. This type of alkene is specifically termed “hyperstable”. There are some problems in the energy balance you have calculated; in order to balance the hydrogenation equation and get a heat of reaction you need to include the other reactant (hydrogen). Also it is not strictly possible to compare energies of non-isomeric compounds (using molecular mechanics) because the more the structure changes, the greater the difference in the scale being used to measure the energy.

Q3. Good effort getting all those vibrations in! The way you have presented the data here is also great - concise and easily referenced to the relevant structures. Your calculations are all fine, but you should include the energy values you got for these structures also (it’s one of the ways in which the calculation can be assessed on marking – and this will probably also be true for later parts of the course). The orbitals look good and the most nucleophilic double bond is correctly predicted. Discussion of the pi-sigma* interaction and its effect on the IR stretches is good.

Q4. Having all of those jmols in the table is making the page a bit buggy – maybe better to use the option of having a link or button for them. Methyl is definitely the most sensible choice for the R group to save computing time (especially for this brief study). Your energy values are pretty close to expected for the PM6 calculations (MM2 a little off – but it’s difficult to comment on why this might be the case without any image or link). I would say that the major reason for the difference in Ca/Cb and Da/Db is not the anomeric effect but the instability of a trans-fused 5-6 ring system over the more common cis-fused – a range of strains are the cause of this difference. You should have found that using MOPAC A=C and B=D; the structures can’t be distinguished because the true structure is something of a hybrid. MM2 won’t give this result because the method can’t determine that there should be new bonds because its bonds are set by the user. The ratio from your Boltzmann actually suggest that there will be 10^16 more of the unfavoured isomer present; the difference in energy is always a positive quantity and the true form of the equation should be the exponent of a negative number; the actual ratio seems massively biased, but I suppose the difference in energy is pretty big in this case.

MP. Your introduction is good – having a concise summary of the key parts of the paper you’re looking at is a good idea. This first question you look at (epimerisation of menthone) is interesting, but I’m not sure exactly what you’re getting at by the end. If a species undergoes epimerisation in this manner, the result is to obtain the thermodynamic product. If you simply left menthone in the presence of the base, over time you would expect to obtain an equilibrium mixture of the + and – isomers through the common enolate form. What is confusing is that you say complete epimerisation occurs but you have the trans isomer to start with (surely complete epimerisation would mean formation of the cis isomer); the 97:3 ratio seems to be about right given the difference in energies you calculated for the two forms. I think I understand what you are looking at next but there are some issues with terminology: these enolate forms are not transition states, they are intermediates. That said, it is reasonable to calculate the energies of these intermediates to see which is more likely to form and explain why the product distribution may not match the starting distribution. What is odd is that you say the cis enolate is lower in energy that should give the cis relationship in the formylated product, but the product you give in the scheme is actually derived from the trans enolate. Did you mean to say that your calculations here do not match the experimental observations (this needs to be made clearer). Incidentally the energy values obtained may be a little off because there is no counter cation included and also it requires quite sophisticated techniques to accurately model ionic species (usually with solvent modelling included). Your NMR and IR results are fine but it would have been better to look at how close the NMR data for the two isomers is and then analyse whether you can differentiate between the two using computational chemistry – i.e. is it possible to say that you definitely have one isomer because the experimental data matches its computed data much better. Also when talking about the comparison of your results to the literature values it would have been worth commenting on how the authors of the paper worked out which isomer they got (2D NMR probably).

Talk:Mod:ms7109 module1

2011-12-02T16:16:41Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your energy values are all spot on although it was not necessary to give data to so many decimal p..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your energy values are all spot on although it was not necessary to give data to so many decimal places: these are not precisely measured quantities and there is also some degree of error due to random fluctuation from one calculation to the next or one laptop to the next. The analysis of strain contributions is good. Thermodynamic vs kinetic control is correctly described for these examples; as you say for the hydrogenated compounds – you know which product would be formed under thermodynamic control, but for kinetically controlled reactions information about the transition states must be obtained. It is good that you have included some references here, but it should be noted that while Baldwin provided an explanation for endo control, the original “determination” that the endo isomer is preferred was made empirically at the turn of the century by Diels and Alder.

Q2. The approach you have taken makes sense (looking at the part of the structure for which various stable conformations are known). Most of your structures and energy values are good. Your calculation for the up-chair isomer does not have the correct structure – the geometry of the double bond is incorrect. It is important to always check the structure after minimisation to ensure it still has the correct configuration (this can easily be compromised when the structure is tweaked to find different starting points). It would have been good to hear about how else you minimised the structure – were there alternative conformations for the central medium ring? Even if this was not the case, it is worth stating that. The strain in hydrogenation is well described. This type of alkene is specifically termed “hyperstable”. There are some problems in the energy balance you have calculated; in order to balance the hydrogenation equation and get a heat of reaction you need to include the other reactant (hydrogen). Also it is not strictly possible to compare energies of non-isomeric compounds (using molecular mechanics) because the more the structure changes, the greater the difference in the scale being used to measure the energy.

Q3. Good effort getting all those vibrations in! The way you have presented the data here is also great - concise and easily referenced to the relevant structures. Your calculations are all fine, but you should include the energy values you got for these structures also (it’s one of the ways in which the calculation can be assessed on marking – and this will probably also be true for later parts of the course). The orbitals look good and the most nucleophilic double bond is correctly predicted. Discussion of the pi-sigma* interaction and its effect on the IR stretches is good.

Q4. Having all of those jmols in the table is making the page a bit buggy – maybe better to use the option of having a link or button for them. Methyl is definitely the most sensible choice for the R group to save computing time (especially for this brief study). Your energy values are pretty close to expected for the PM6 calculations (MM2 a little off – but it’s difficult to comment on why this might be the case without any image or link). I would say that the major reason for the difference in Ca/Cb and Da/Db is not the anomeric effect but the instability of a trans-fused 5-6 ring system over the more common cis-fused – a range of strains are the cause of this difference. You should have found that using MOPAC A=C and B=D; the structures can’t be distinguished because the true structure is something of a hybrid. MM2 won’t give this result because the method can’t determine that there should be new bonds because its bonds are set by the user. The ratio from your Boltzmann actually suggest that there will be 10^16 more of the unfavoured isomer present; the difference in energy is always a positive quantity and the true form of the equation should be the exponent of a negative number; the actual ratio seems massively biased, but I suppose the difference in energy is pretty big in this case.

MP. Your introduction is good – having a concise summary of the key parts of the paper you’re looking at is a good idea. This first question you look at (epimerisation of menthone) is interesting, but I’m not sure exactly what you’re getting at by the end. If a species undergoes epimerisation in this manner, the result is to obtain the thermodynamic product. If you simply left menthone in the presence of the base, over time you would expect to obtain an equilibrium mixture of the + and – isomers through the common enolate form. What is confusing is that you say complete epimerisation occurs but you have the trans isomer to start with (surely complete epimerisation would mean formation of the cis isomer); the 97:3 ratio seems to be about right given the difference in energies you calculated for the two forms. I think I understand what you are looking at next but there are some issues with terminology: these enolate forms are not transition states, they are intermediates. That said, it is reasonable to calculate the energies of these intermediates to see which is more likely to form and explain why the product distribution may not match the starting distribution. What is odd is that you say the cis enolate is lower in energy that should give the cis relationship in the formylated product, but the product you give in the scheme is actually derived from the trans enolate. Did you mean to say that your calculations here do not match the experimental observations (this needs to be made clearer). Incidentally the energy values obtained may be a little off because there is no counter cation included and also it requires quite sophisticated techniques to accurately model ionic species (usually with solvent modelling included). Your NMR and IR results are fine but it would have been better to look at how close the NMR data for the two isomers is and then analyse whether you can differentiate between the two using computational chemistry – i.e. is it possible to say that you definitely have one isomer because the experimental data matches its computed data much better. Also when talking about the comparison of your results to the literature values it would have been worth commenting on how the authors of the paper worked out which isomer they got (2D NMR probably).

Talk:Mod:1992

2011-12-02T16:16:10Z

Mjhughes:

Q1: Your energy values are all good but it isn’t necessary to give so many decimal places – there is some error to take into account in these approximated quantities. The analysis of strain contributions is well done – the torsional strain is strictly determined by deviation from ideal dihedral angle; this is affected by sterics (as you discussed) but also by the possibility for hyperconjugation and corresponding stabilisation. Also you referred to the alkenic carbons in the hydrogenated compounds as sp3 – I think this was just a typo but be careful in getting important terms correct when editing. All of the discussion of thermodynamic vs kinetic control is correct as are the conclusions you can make about the reactions with the limited information you have.

Q2. There are some problems with your structures for these compounds. In each case the double bond geometry and ring fusion geometry are incorrect – should have the hydrogen of the double bond cis to the gem-dimethyl bridging group and the ring junction should be cis fused (hydrogens on the same side). It is very important to double check the structures you are working on to make sure they still match the compound in question. In answering this question it would have been good to see some different conformations (not necessarily just the low energy ones) and a description of the steps taken in performing the minimisation since this is the bulk of the work required. The definition of hyperstable alkenes is correct and the qualitative analysis of the structure of the hydrogenated derivative is good. You correctly recognised that even if you cannot compare the energy values (using MM2) you can compare the structures generated.

Q3. Some of the MOs look good, although it looks to me that the HOMO has more density on the side anti to Cl. The results of this these surface calculations can be very inconsistent and sometimes it is worth running them a few times with different minimisation attempts (and also symmetrisation) to get them looking more symmetrical and with the appropriate lobes. The IR stretches are all as expected and indeed the C-Cl bond is weakened when the anti-double bond is present due to the pi-sigma* interaction. One side note – you should include all of the energy values you get somewhere (e.g. those from the MOPAC methods) because these are used in part to assess the calculation; this probably applies to the other parts of the course too.

Q4. R=methyl is definitely the right choice for this question. Your energies are all a little off because you have missed an oxygen atom out of the structure the reactive acetyl group should be bound to the ring by this missing oxygen (e.g. OAc not CH2Ac). Aside from this the structures look about as expected. As your results show (both structures and energies) A=C and B=D when MOPAC methods are used. This is because MOPAC can account for the attack of the acetyl group (bond formation); in contrast, MM2 cannot imagine bonds to be any different to the initial input. You are right to say that the C’ and D’ forms are so high in energy they are present in vanishingly small quantities and do not influence the overall stereoselectivity; the approach for nucleophilic attack is also better in C/D.

MP. This NMR data looks reasonably close to the lit values – I think the sets of data would be better presented in a table for this question to clearly see the deviations (or better still graphically in a bar chart for example). You could have analysed this a bit more quantitatively by finding the average error. Working out ratios of peaks in 13C NMR is not normally done because peak intensity does not correspond to relative abundance in the same way as in 1H NMR. This is because the relaxation time of the 13C nucleus is slower and not all nuclei will have been relaxed in time in between scans. Usually the carbon NMR is analysed by counting the number of peaks to ensure that the structure is correct and equivalent nuclei will be taken into account. It is perfectly logical to average out the resonances you get for peaks made non-equivalent in the static form, because that is exactly what is happening experimentally with the rotation of the bond averaging out the chemical environments. You calculated NMR spectra for both forms of the product which is exactly the right approach to see whether you can differentiate between the two forms – to follow up on this you could have shown how close each of the computed resonances match the experimental data to see to what extent the different isomers can be differentiated.

Talk:Mod:1992

2011-12-02T16:15:41Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your energy values are all good but it isn’t necessary to give so many decimal places – there i..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your energy values are all good but it isn’t necessary to give so many decimal places – there is some error to take into account in these approximated quantities. The analysis of strain contributions is well done – the torsional strain is strictly determined by deviation from ideal dihedral angle; this is affected by sterics (as you discussed) but also by the possibility for hyperconjugation and corresponding stabilisation. Also you referred to the alkenic carbons in the hydrogenated compounds as sp3 – I think this was just a typo but be careful in getting important terms correct when editing. All of the discussion of thermodynamic vs kinetic control is correct as are the conclusions you can make about the reactions with the limited information you have.

Q2. There are some problems with your structures for these compounds. In each case the double bond geometry and ring fusion geometry are incorrect – should have the hydrogen of the double bond cis to the gem-dimethyl bridging group and the ring junction should be cis fused (hydrogens on the same side). It is very important to double check the structures you are working on to make sure they still match the compound in question. In answering this question it would have been good to see some different conformations (not necessarily just the low energy ones) and a description of the steps taken in performing the minimisation since this is the bulk of the work required. The definition of hyperstable alkenes is correct and the qualitative analysis of the structure of the hydrogenated derivative is good. You correctly recognised that even if you cannot compare the energy values (using MM2) you can compare the structures generated.

Q3. Some of the MOs look good, although it looks to me that the HOMO has more density on the side anti to Cl. The results of this these surface calculations can be very inconsistent and sometimes it is worth running them a few times with different minimisation attempts (and also symmetrisation) to get them looking more symmetrical and with the appropriate lobes. The IR stretches are all as expected and indeed the C-Cl bond is weakened when the anti-double bond is present due to the pi-sigma* interaction. One side note – you should include all of the energy values you get somewhere (e.g. those from the MOPAC methods) because these are used in part to assess the calculation; this probably applies to the other parts of the course too.

Q4. R=methyl is definitely the right choice for this question. Your energies are all a little off because you have missed an oxygen atom out of the structure the reactive acetyl group should be bound to the ring by this missing oxygen (e.g. OAc not CH2Ac). Aside from this the structures look about as expected. As your results show (both structures and energies) A=C and B=D when MOPAC methods are used. This is because MOPAC can account for the attack of the acetyl group (bond formation); in contrast, MM2 cannot imagine bonds to be any different to the initial input. You are right to say that the C’ and D’ forms are so high in energy they are present in vanishingly small quantities and do not influence the overall stereoselectivity; the approach for nucleophilic attack is also better in C/D.

MP. This NMR data looks reasonably close to the lit values – I think the sets of data would be better presented in a table for this question to clearly see the deviations (or better still graphically in a bar chart for example). You could have analysed this a bit more quantitatively by finding the average error. Working out ratios of peaks in 13C NMR is not normally done because peak intensity does not correspond to relative abundance in the same way as in 1H NMR. This is because the relaxation time of the 13C nucleus is slower and not all nuclei will have been relaxed in time in between scans. Usually the carbon NMR is analysed by counting the number of peaks to ensure that the structure is correct and equivalent nuclei will be taken into account. It is perfectly logical to average out the resonances you get for peaks made non-equivalent in the static form, because that is exactly what is happening experimentally with the rotation of the bond averaging out the chemical environments. You calculated NMR spectra for both forms of the product which is exactly the right approach to see whether you can differentiate between the two forms – to follow up on this you could have shown how close each of the computed resonances match the experimental data to see to what extent the different isomers can be differentiated.

Talk:Mod:BTL091

2011-12-02T16:15:16Z

Mjhughes:

Q1: Your energy values are all correct, but it is not necessary to give so many decimal places; there is some error in these quantities because they are approximations and the program will show some variation from one calculation to the next. The strain contributions in the hydrogenated compounds is well analysed. Discussion of thermodynamic vs kinetic control is mostly good; in the case of the hydrogenation, you can only say for sure that compound 4 is favoured in the process if it is under thermodynamic control. If the reaction is under kinetic control however, more information (and DFT calculation) would be needed to determine which pathway has the lowest energy activation energy.

Q2. The structures given have the wrong geometry at the ring junction (should be cis-fused). It is very important to regularly check that the structure you are working on matches the compound in question. It would have been good to see some more examples of different conformational isomers (like you did for 10a and 10b) to discuss how you minimised the structure. For example you said you only looked at conformations with the 6-ring in the chair conformation but you could have shown the twist-boat conformations as well. Although cyclohexane prefers a chair-conformation, this is not always the case for substituted 6-rings. The definition of hyperstable alkenes is good.

Q3. Your MOs look good and the explanation for the regioselectivity is correct. The IR stretches are good for the alkenes; the ones for the C-Cl bonds are a little low (possibly C-H vibrations? – it is difficult to tell without an image of the vibration). The C-Cl stretch of the monohydrogenated compound should be higher in value than for the dialkene – your result contradicts this although you say differently! (Wavenumbers are proportional to energy and hence bond strength). If your conclusion (or the lit result) doesn’t add up to your calculated result there has to be some explanation given or maybe it suggests you need to look back at your calculation.

Q4. R=methyl is indeed the best model to use here. Your energy values and structures are pretty much exactly right! You are right to say that the selectivity is largely governed by the huge difference in energy between C/D and C’/D’ giving the higher energy configurations only in vanishingly small quantities. There is also a kinetic factor - the trajectory for nucleophilic attack is better in C/D than C’/D’. The major difference between MM2 calculations and MOPAC calculations here is that MM2 cannot incorporate the neighbouring group participation whereas MOPAC can. You can see from your energy values and your structures that A=C and B=D. This is because with the carbonyl oxygen proximal to the oxonium carbon, the method recognises that bond formation is possible (this isn’t possible in MM2 because the bonds are set at the start).

MP. Wrt regioselectivity: More specifically, the selectivity depends on the ability of the migrating group to stabilise positive charge because this is the group that develops the most charge in the transition state. You’re calculated NMR data looks like a reasonable fit. It makes more sense to discuss the difference in ppm rather than the % difference in this case when analysing the error here and it the deviations are often depicted graphically (in a bar chart for example). The problem with using 15N NMR to attempt to distinguish between the isomers is that you’d have to make the 15N enriched product typically introducing N as ammonia or another cheap feedstock. IR is about as accurate as is expect – not easy to model especially since the calculation is done in the gas phase (experimentally IR is measured in condensed phases). You are right to mention inaccuracy possibly arising from different conformational preferences of the compounds and the difficulty in obtaining a global minimum – to elaborate on this it would be necessary to analyse different conformations available (as in Q2).

Talk:Mod:BTL091

2011-12-02T16:14:56Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your energy values are all correct, but it is not necessary to give so many decimal places; there ..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your energy values are all correct, but it is not necessary to give so many decimal places; there is some error in these quantities because they are approximations and the program will show some variation from one calculation to the next. The strain contributions in the hydrogenated compounds is well analysed. Discussion of thermodynamic vs kinetic control is mostly good; in the case of the hydrogenation, you can only say for sure that compound 4 is favoured in the process if it is under thermodynamic control. If the reaction is under kinetic control however, more information (and DFT calculation) would be needed to determine which pathway has the lowest energy activation energy.

Q2. The structures given have the wrong geometry at the ring junction (should be cis-fused). It is very important to regularly check that the structure you are working on matches the compound in question. It would have been good to see some more examples of different conformational isomers (like you did for 10a and 10b) to discuss how you minimised the structure. For example you said you only looked at conformations with the 6-ring in the chair conformation but you could have shown the twist-boat conformations as well. Although cyclohexane prefers a chair-conformation, this is not always the case for substituted 6-rings. The definition of hyperstable alkenes is good.

Q3. Your MOs look good and the explanation for the regioselectivity is correct. The IR stretches are good for the alkenes; the ones for the C-Cl bonds are a little low (possibly C-H vibrations? – it is difficult to tell without an image of the vibration). The C-Cl stretch of the monohydrogenated compound should be higher in value than for the dialkene – your result contradicts this although you say differently! (Wavenumbers are proportional to energy and hence bond strength). If your conclusion (or the lit result) doesn’t add up to your calculated result there has to be some explanation given or maybe it suggests you need to look back at your calculation.

Q4. R=methyl is indeed the best model to use here. Your energy values and structures are pretty much exactly right! You are right to say that the selectivity is largely governed by the huge difference in energy between C/D and C’/D’ giving the higher energy configurations only in vanishingly small quantities. There is also a kinetic factor - the trajectory for nucleophilic attack is better in C/D than C’/D’. The major difference between MM2 calculations and MOPAC calculations here is that MM2 cannot incorporate the neighbouring group participation whereas MOPAC can. You can see from your energy values and your structures that A=C and B=D. This is because with the carbonyl oxygen proximal to the oxonium carbon, the method recognises that bond formation is possible (this isn’t possible in MM2 because the bonds are set at the start).

MP. Wrt regioselectivity: More specifically, the selectivity depends on the ability of the migrating group to stabilise positive charge because this is the group that develops the most charge in the transition state. You’re calculated NMR data looks like a reasonable fit. It makes more sense to discuss the difference in ppm rather than the % difference in this case when analysing the error here and it the deviations are often depicted graphically (in a bar chart for example). The problem with using 15N NMR to attempt to distinguish between the isomers is that you’d have to make the 15N enriched product typically introducing N as ammonia or another cheap feedstock. IR is about as accurate as is expect – not easy to model especially since the calculation is done in the gas phase (experimentally IR is measured in condensed phases). You are right to mention inaccuracy possibly arising from different conformational preferences of the compounds and the difficulty in obtaining a global minimum – to elaborate on this it would be necessary to analyse different conformations available (as in Q2).

Talk:Mod:SergioGeorgini

2011-12-02T16:14:26Z

Mjhughes:

Please make sure you put in references before submission, particularly for Figure 4.
====Q1====
*Torsion energy explanation: dihedral angles.
*Your analysis on the source of the difference in bend energies is correct, but Figure 6 and 7 are hard to see until I click on the picture. Tip: remove empty space.
====Q2====
*You seem to be fixed on 10 being more stable at the beginning even before showing the calculation results.
*"lowest totsl energy"
*Table 4: you need to clarify where the breakdown of energies come from. Are there any significant difference using MMFF94?
*You seem to understand the point I raised above with your comparison. However, the more corrected comparison between different techniques is the absolute difference in energy between 10 and 9. This is an observable quantity, which is directly related to the equilibrium constant between 9 and 10. Do MM2 and MMFF94 agree with each other?
*You were spot-on with hyperstable alkenes and the comparison in energy with hydrogenated product is appropriate. However, if you want to compare the energies of two different systems, you must BALANCE THE EQUATION.
====Q3====
*You might want to clearly explain how the structures were optimised and how the MOs were calculated. Instructions are included in the question, but we would still want to be absolutely certain which techniques students used. Sometime you guys even do a better job than we prescribed.
*Figure 14 -15: the main gripe is that there is too much empty space and the molecules I can see are consequently small. Clicking on the picture helps, but if you can please trim the pictures.
*Your MOs could be made smoother and nicer using Chem3D options. Talk to us if you don't know how.
*You talked about PM6 at the beginning, and then PM3 at the end. Please double check!
*Correct spotting of the ?*-? interaction, but you'll need to provide a better explanation for the selectivity. What is the carbene looking for to react? What's the mechanism?
====Q4====
*A' picture missing.
*A' MM2 and PM6 jmol missing. I do like it that you included the jmols from both methods.
*Please double check D and D'.
*You correctly spotted that energies of A and C are the same using PM6. Are the structures the same? and if MM2 gives different results, which method is wrong?
*From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?
====Overall====
I think you did a good job for week 1. Obviously more careful analysis and tuning to data presentation is required, but most of the data is solid. If you still aren't sure about something, do come and talk with us.

====After Marking====

Intro: Most of your comments here are good (e.g. the description of the different calculation methods). While it is fair to say that finding a very high energy transition state would suggest this is an unlikely pathway, this is not really the way in which synthetic chemists typically use computational chemistry. It breaks down more like this: There is a simple approach which is to embellish a report on a new reaction or facet of chemistry by calculating a feasible transition state (rather than just speculating with pen and paper mechanisms). A better approach is to compare energies found to empirical results (e.g. comparison of difference in calculated activation energy with observed ee) to add further weight to the model. The most sophisticated option (more like what you suggest) is to analyse computationally a wide range of different conditions (e.g. different catalysts in a known reaction mechanism) and then try out the system that gives the best results – this is usually done if carrying out all of the conditions is difficult (for example if catalysts are hard to make); it wouldn’t be done for example to find the best solvent (which is a cheap screen of conditions to perform).

Q1: Your energy values are all spot on – although it is not necessary to include so many decimal places in the results; consider that MM2 is an approximate method and there is also some random error in the calculation. Contributions to strain are correctly identified and discussed well. Using MM2 methods the more the structure starts to change (i.e. going from monounsaturated to saturated) the more inaccurate it becomes to compare the energies calculated – effectively the scale on which the energy values are measured changes. So overall, while your approach to calculating the energy balance for a second hydrogenation is correct, it would have been better to use a DFT calculation to get the energy values. What is still reasonable is the qualitative analysis you have given on the structure of the saturated product which suggest that it will be highly strained. The discussion of kinetic vs thermodynamic control is all correct.

Q2. Your approach to this question is good – analyse some different conformations and focus on features with well defined possibilities (6-ring conformations), but there are some problems with you structures: Your chair conformations have the wrong geometry at the ring junction (should be cis fused – i.e. hydrogens on the same side). It is very important to keep checking the structure you are working on because it can easily change from the correct form on manipulation. The discussion of hyperstable alkenes is very good. It is fine to make qualitative comparisons of the strain in the alkene and parent hydrocarbon as you have done. The comparison of energies is more problematic (with molecular mechanics) because the more a structure changes the more different is the scale on which the energy is calculated – you have the right approach though, including hydrogen in the energy balance; it would be interesting to see how the MM results compare to some calculated by DFT methods.

Q3. Your testing of different MOPAC methods to get more accurate, symmetrical MOs was a good step to take. The MOs look good and the explanation about the reactivity is all correct. The stretching frequencies are also correct as is the explanation for the differences that arise due to the pi-sigma* interaction. The only other comment I have to make about this question is that you should include all of the energies you get from different methods (e.g. MOPAC); this is useful because it can be used to assess the accuracy of your structures/energies.

Q4. R=methyl is indeed the best model to use here. Your energy values and structures are very good. As you recognised MOPAC gives A=C etc because the method can determine the bonding character between the carbonyl oxygen and the oxonium carbon when they are proximal; MM2 can’t do this because the bonds are set by the user at the start of the calculation. The selectivity is indeed largely governed by the relative distributions of C/C’ and D/D’; another factor is that the trajectory for nucleophilic attack is also better in the lower energy conformations.

MP. The introduction is good – setting out the questions you want to answer. Your NMR data looks fine and the approach in analysing your results is right – comparing two sets of calculated data to the lit values tells you whether you can differentiate between the isomers; this is often presented graphically (e.g. in a bar chart). The IR is also about as accurate as can be expected. I doubt that optical rotation was what the authors of the paper actually used to differentiate between the species – it would be necessary to already know which isomer has which rotation and compare the experimental result. I imagine they performed some additional 2D NMR experiments to determine which isomer they had – these type of experiments provide information about which Hs are close in space and which are a certain number of bonds apart. For the discussion of reactivity in the methylation reaction – it is reasonable to use this approach of looking at the structure of the starting material to determine possible factors affecting which side will be attacked. One thing to consider is whether the nucleophile in this case is actually bigger than the hydroxy group being formed (if the carbonyl oxygen is coordinated by Ce it will become even bigger). When the 1,2 addition takes place, the oxygen atom will be forced in one direction or the other, so the effect of this also needs to be considered. Overall, it is best to use a higher level of computation and look into transition state modelling to look at these issues in more depth (for example it can be difficult to predict the effect of dipole interactions and other non-steric factors).

Talk:Mod:SergioGeorgini

2011-12-02T16:14:05Z

Mjhughes:

Talk:Mod:SergioGeorgini

2011-12-02T16:13:05Z

Mjhughes:

Talk:Mod:SergioGeorgini

2011-12-02T16:12:40Z

Mjhughes:

Please make sure you put in references before submission, particularly for Figure 4.
====Q1====
*Torsion energy explanation: dihedral angles.
*Your analysis on the source of the difference in bend energies is correct, but Figure 6 and 7 are hard to see until I click on the picture. Tip: remove empty space.
====Q2====
*You seem to be fixed on 10 being more stable at the beginning even before showing the calculation results.
*"lowest totsl energy"
*Table 4: you need to clarify where the breakdown of energies come from. Are there any significant difference using MMFF94?
*You seem to understand the point I raised above with your comparison. However, the more corrected comparison between different techniques is the absolute difference in energy between 10 and 9. This is an observable quantity, which is directly related to the equilibrium constant between 9 and 10. Do MM2 and MMFF94 agree with each other?
*You were spot-on with hyperstable alkenes and the comparison in energy with hydrogenated product is appropriate. However, if you want to compare the energies of two different systems, you must BALANCE THE EQUATION.
====Q3====
*You might want to clearly explain how the structures were optimised and how the MOs were calculated. Instructions are included in the question, but we would still want to be absolutely certain which techniques students used. Sometime you guys even do a better job than we prescribed.
*Figure 14 -15: the main gripe is that there is too much empty space and the molecules I can see are consequently small. Clicking on the picture helps, but if you can please trim the pictures.
*Your MOs could be made smoother and nicer using Chem3D options. Talk to us if you don't know how.
*You talked about PM6 at the beginning, and then PM3 at the end. Please double check!
*Correct spotting of the ?*-? interaction, but you'll need to provide a better explanation for the selectivity. What is the carbene looking for to react? What's the mechanism?
====Q4====
*A' picture missing.
*A' MM2 and PM6 jmol missing. I do like it that you included the jmols from both methods.
*Please double check D and D'.
*You correctly spotted that energies of A and C are the same using PM6. Are the structures the same? and if MM2 gives different results, which method is wrong?
*From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?
====Overall====
I think you did a good job for week 1. Obviously more careful analysis and tuning to data presentation is required, but most of the data is solid. If you still aren't sure about something, do come and talk with us.

Normal 0 false false false EN-GB X-NONE X-NONE

Intro: Most of your comments here are good (e.g. the description of the different calculation methods). While it is fair to say that finding a very high energy transition state would suggest this is an unlikely pathway, this is not really the way in which synthetic chemists typically use computational chemistry. It breaks down more like this: There is a simple approach which is to embellish a report on a new reaction or facet of chemistry by calculating a feasible transition state (rather than just speculating with pen and paper mechanisms). A better approach is to compare energies found to empirical results (e.g. comparison of difference in calculated activation energy with observed ee) to add further weight to the model. The most sophisticated option (more like what you suggest) is to analyse computationally a wide range of different conditions (e.g. different catalysts in a known reaction mechanism) and then try out the system that gives the best results – this is usually done if carrying out all of the conditions is difficult (for example if catalysts are hard to make); it wouldn’t be done for example to find the best solvent (which is a cheap screen of conditions to perform).

Q1: Your energy values are all spot on – although it is not necessary to include so many decimal places in the results; consider that MM2 is an approximate method and there is also some random error in the calculation. Contributions to strain are correctly identified and discussed well. Using MM2 methods the more the structure starts to change (i.e. going from monounsaturated to saturated) the more inaccurate it becomes to compare the energies calculated – effectively the scale on which the energy values are measured changes. So overall, while your approach to calculating the energy balance for a second hydrogenation is correct, it would have been better to use a DFT calculation to get the energy values. What is still reasonable is the qualitative analysis you have given on the structure of the saturated product which suggest that it will be highly strained. The discussion of kinetic vs thermodynamic control is all correct.

Q2. Your approach to this question is good – analyse some different conformations and focus on features with well defined possibilities (6-ring conformations), but there are some problems with you structures: Your chair conformations have the wrong geometry at the ring junction (should be cis fused – i.e. hydrogens on the same side). It is very important to keep checking the structure you are working on because it can easily change from the correct form on manipulation. The discussion of hyperstable alkenes is very good. It is fine to make qualitative comparisons of the strain in the alkene and parent hydrocarbon as you have done. The comparison of energies is more problematic (with molecular mechanics) because the more a structure changes the more different is the scale on which the energy is calculated – you have the right approach though, including hydrogen in the energy balance; it would be interesting to see how the MM results compare to some calculated by DFT methods.

Q3. Your testing of different MOPAC methods to get more accurate, symmetrical MOs was a good step to take. The MOs look good and the explanation about the reactivity is all correct. The stretching frequencies are also correct as is the explanation for the differences that arise due to the pi-sigma* interaction. The only other comment I have to make about this question is that you should include all of the energies you get from different methods (e.g. MOPAC); this is useful because it can be used to assess the accuracy of your structures/energies.

Q4. R=methyl is indeed the best model to use here. Your energy values and structures are very good. As you recognised MOPAC gives A=C etc because the method can determine the bonding character between the carbonyl oxygen and the oxonium carbon when they are proximal; MM2 can’t do this because the bonds are set by the user at the start of the calculation. The selectivity is indeed largely governed by the relative distributions of C/C’ and D/D’; another factor is that the trajectory for nucleophilic attack is also better in the lower energy conformations.

MP. The introduction is good – setting out the questions you want to answer. Your NMR data looks fine and the approach in analysing your results is right – comparing two sets of calculated data to the lit values tells you whether you can differentiate between the isomers; this is often presented graphically (e.g. in a bar chart). The IR is also about as accurate as can be expected. I doubt that optical rotation was what the authors of the paper actually used to differentiate between the species – it would be necessary to already know which isomer has which rotation and compare the experimental result. I imagine they performed some additional 2D NMR experiments to determine which isomer they had – these type of experiments provide information about which Hs are close in space and which are a certain number of bonds apart. For the discussion of reactivity in the methylation reaction – it is reasonable to use this approach of looking at the structure of the starting material to determine possible factors affecting which side will be attacked. One thing to consider is whether the nucleophile in this case is actually bigger than the hydroxy group being formed (if the carbonyl oxygen is coordinated by Ce it will become even bigger). When the 1,2 addition takes place, the oxygen atom will be forced in one direction or the other, so the effect of this also needs to be considered. Overall, it is best to use a higher level of computation and look into transition state modelling to look at these issues in more depth (for example it can be difficult to predict the effect of dipole interactions and other non-steric factors).

Talk:Mod:sp31

2011-12-02T16:11:50Z

Mjhughes:

Q1: Your energy values are spot on – but you have the wrong structure for compound 3 (it should be derived from the endo-product as shown in your Figure 1). The major difference between 1 and 2 is indeed torsional strain (deviation from ideal dihedral angles); for 3 and 4 it is bending strain due to differences in deviation from ideal sp2 bond angles – you still found that to be the case with the wrong structure. Another point to take care on is the naming of these compounds: it would be fine to label them all just compound or molecule 1, 2 etc...None of these structures can be described as “cyclopentadiene” however. The discussion about kinetic vs thermodynamic control is fine. The endo-dimer is shown to be a kinetic product because it is formed in spite of it not being the most stable product. It is not true to that the kinetic product is by definition the one that is higher in energy (i.e. it is not some kind of opposite to the thermodynamic product). The kinetic product is simply the one that comes from the lowest energy transition state and in fact the kinetic product and thermodynamic product is often the same. I wasn’t entirely sure if this had been understood (the relevant statement could be interpreted either way) so this is here to clear this up in case. As for the hydrogenation it is fine to say the lower energy compound is expected to form if you include the qualifier “under thermodynamic control”.

Q2. Your structures have the correct configuration but there is a lower energy conformation available for intermediate 1. It would have been worth discussing how you came to these minimised structures – e.g. show some other higher energy conformations. A key feature that is a good starting point here is the 6-ring; there are well-known conformations available for such structures and it is worth analysing some of them (chair vs twist-boat). In your structure for int 1 the ring is in the twist-boat when it should be chair. You mention why the structures have different energies which is good, but a fuller answer could have compared the strain contributions as in Q1. Part of the course is training in the techniques – so if you see an opportunity to apply something that was used in an earlier question go ahead because you’ll get extra marks for it. The definition of hyperstable alkenes is good.

Q3. Your MOs are good and the rationalisation for the regioselectivity is correct. The IR stretches you have listed are a little off in some cases – the C-Cl bond is expected to be strengthened when the double bond anti to it is removed, this is because there is a pi-sigma* interaction between it and the C-Cl antibonding orbital (LUMO+1). This is also the reason for the difference in the stretching frequencies of the two alkenes in the initial compound. Without seeing the actual stretches it is hard to say whether this is because the wrong peaks are listed – it is worth including even some images that illustrate the vibrations so that they could be checked.

Q4. R=Me is the right choice. Your PM6 energies are all pretty close and the structures look good. For some of them (B for example) a lower energy conformation is accessible if the carbonyl oxygen is positioned directly above the oxonium. This may not be the case in the MM2 calculations but the difference with MOPAC is the neighbouring group participation can be taken into account directly. With the oxygen and the oxonium close together it will recognise the possibility for bonding and the structure for A will be the same as C (likewise B=D). As you say the major deciding factor for the selectivity is the C/C’ or D/D’ ratio which is heavily biased towards the cis-fused structures.

MP. You have taken a good approach here in comparing both sets of calculated data to the experimental data. As you have shown it doesn’t appear possible to distinguish these isomers using computational chemistry. It can be difficult to calculate the NMR spectrum for cyclopropyl compounds because they are highly strained (as you said) and the associated MOs are unusual. For the phosphorus NMR, the inaccuracy could be due to conformational flexibility of the phosphorus substituents. 31P NMR can sometimes be hard to predict even experimentally, so this problem could require application of different computational methods. Other things to look at would have been IR and optical rotation which in principle could be different for the different isomers and the difference in energy and its significance to the reaction mechanism.

Talk:Mod:sp31

2011-12-02T16:11:27Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your energy values are spot on – but you have the wrong structure for compound 3 (it should be de..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your energy values are spot on – but you have the wrong structure for compound 3 (it should be derived from the endo-product as shown in your Figure 1). The major difference between 1 and 2 is indeed torsional strain (deviation from ideal dihedral angles); for 3 and 4 it is bending strain due to differences in deviation from ideal sp2 bond angles – you still found that to be the case with the wrong structure. Another point to take care on is the naming of these compounds: it would be fine to label them all just compound or molecule 1, 2 etc...None of these structures can be described as “cyclopentadiene” however. The discussion about kinetic vs thermodynamic control is fine. The endo-dimer is shown to be a kinetic product because it is formed in spite of it not being the most stable product. It is not true to that the kinetic product is by definition the one that is higher in energy (i.e. it is not some kind of opposite to the thermodynamic product). The kinetic product is simply the one that comes from the lowest energy transition state and in fact the kinetic product and thermodynamic product is often the same. I wasn’t entirely sure if this had been understood (the relevant statement could be interpreted either way) so this is here to clear this up in case. As for the hydrogenation it is fine to say the lower energy compound is expected to form if you include the qualifier “under thermodynamic control”.

Q2. Your structures have the correct configuration but there is a lower energy conformation available for intermediate 1. It would have been worth discussing how you came to these minimised structures – e.g. show some other higher energy conformations. A key feature that is a good starting point here is the 6-ring; there are well-known conformations available for such structures and it is worth analysing some of them (chair vs twist-boat). In your structure for int 1 the ring is in the twist-boat when it should be chair. You mention why the structures have different energies which is good, but a fuller answer could have compared the strain contributions as in Q1. Part of the course is training in the techniques – so if you see an opportunity to apply something that was used in an earlier question go ahead because you’ll get extra marks for it. The definition of hyperstable alkenes is good.

Q3. Your MOs are good and the rationalisation for the regioselectivity is correct. The IR stretches you have listed are a little off in some cases – the C-Cl bond is expected to be strengthened when the double bond anti to it is removed, this is because there is a pi-sigma* interaction between it and the C-Cl antibonding orbital (LUMO+1). This is also the reason for the difference in the stretching frequencies of the two alkenes in the initial compound. Without seeing the actual stretches it is hard to say whether this is because the wrong peaks are listed – it is worth including even some images that illustrate the vibrations so that they could be checked.

Q4. R=Me is the right choice. Your PM6 energies are all pretty close and the structures look good. For some of them (B for example) a lower energy conformation is accessible if the carbonyl oxygen is positioned directly above the oxonium. This may not be the case in the MM2 calculations but the difference with MOPAC is the neighbouring group participation can be taken into account directly. With the oxygen and the oxonium close together it will recognise the possibility for bonding and the structure for A will be the same as C (likewise B=D). As you say the major deciding factor for the selectivity is the C/C’ or D/D’ ratio which is heavily biased towards the cis-fused structures.

MP. You have taken a good approach here in comparing both sets of calculated data to the experimental data. As you have shown it doesn’t appear possible to distinguish these isomers using computational chemistry. It can be difficult to calculate the NMR spectrum for cyclopropyl compounds because they are highly strained (as you said) and the associated MOs are unusual. For the phosphorus NMR, the inaccuracy could be due to conformational flexibility of the phosphorus substituents. 31P NMR can sometimes be hard to predict even experimentally, so this problem could require application of different computational methods. Other things to look at would have been IR and optical rotation which in principle could be different for the different isomers and the difference in energy and its significance to the reaction mechanism.

Talk:Mod:SAMROWE001

2011-12-02T16:10:48Z

Mjhughes:

====Q1====
*Correct calculation for the dimers, but we're looking for some analysis as well. Where do you think the difference in energy come from? Which part of the molecules is responsible for it? The two hydrogens you mentioned is part of it, but there's more. You'll need to go back to the meaning of each component of the overall energy.
*Again, which part of the monohydrogenated compounds is responsible for the difference in bend energy?
====Q2====
*I appreciate that you discussed the twisted-boat conformers and explained how you arrived at the final optimised structures.
A bit more analysis on the MMFF94 data and components would be great here. Does it give the same structure as MM2?
*I think you 'get' the concept of hyperstable alkenes, but will need to explain it more clearly. Why didn't you show more calculation results of the hydrogenated compound? Can you compare that with the alkene (BIG CLUE: balance the equation!).
====Q3====
*Your picture size is good, but from this point of view I can't see the other half of the molecule. This is important when looking at MOs, so please adjust the pictures.
*A bit more elaboration on the σ*-π interaction (specifying each orbitals are involved and the consequences which you will observe later with your vibrational frequencies calculation) would be great.

====Q4====
*Which complex interactions are we talking about? and why can't MM2 handle them?
*From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?
*Your MM2 structures (which I can't find) aren't the lowest in energy, although your PM6 energies are close (not quite yet). However, it's the analysis and explanation that matter here.
*It would be very important to have a figure showing which structures are A/A' B/B' C/C' and D/D'. I already think you're using a different system from the one in the question.
*" it is able to detect that ring B' shows many unfavourable orbital interactions and therefore distorts the entire geometry to produce a structure which is very similar to that of ring B. " I don't think there is unfavourable interaction here. B is just so more stable. Did you start off with structure B from MM2? Again, since I'm not seeing the jmol, it's all guess work.
*The whole purpose of this question is to let students find out the origin of the diastereoselectivity observed for this kind of reaction. From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?

====Overall====
I think you made a good start, but the analysis side is still lacking for really good mark. Nevertheless, I hope the feeback help and would welcome any further discussion.

====After Marking:====

Q1: Your energy values are all correct and the analysis of the contributions to strain is broadly correct. For compounds 3 and 4 the main difference is in the alkene bond angles rather than the bond angles specifically within the bridged bicyclic part. What you say about kinetic and thermodynamic control is correct – you can’t really say much more about the hydrogenation than you have without more information about the reaction conditions and knowledge of the transition state energies (which would require DFT methods).

Q2. Your energy values and the structures of your global minima are correct. Although you have the right answer it would have been good to see how you got there – e.g. showing some higher energy conformations you found along the way. For example, although 6-rings often adopt chair conformations this is not always the most stable form (sometimes twist-boat or boat can be formed). Bending strain specifically refers to deviation from ideal bond angles so there will be some angles in isomer 9 that deviate more than in isomer 10. The definition of hyperstable alkenes is good and the qualitative analysis of strain in the hydrogenated form is a nice extra – it is not really possible to compare energy values using MM methods because these are not isomeric species (also you need to include the energy of hydrogen to get a balanced equation).

Q3. This question was well answered - the MOs and stretches are correct and both the regioselectivity and the pi-sigma* interaction are fully explained. The only thing to mention is that it is worth inlcuding the energy values you get for all calculations as these are sometimes used to evaluate the calculations.

Q4. R=Me is the correct choice and the energy values you have using PM6 are pretty good; Some of the MM2 results are a bit off (there should still be a significant difference between A/B and A’/B’ (>10kcal/mol) ; without the structures it is hard to see why your energies are different – there is quite a degree of flexibility even in this relatively small system. As you have found, when using MOPAC methods A=C etc. You’re explanation for this is correct – MOPAC can determine bonding interactions if the appropriate atoms are close, whereas MM cannot. NB: The Burgi-Dunitz angle is 107.

MP. Your calculated spectral data all seems to give a good match to the reported values – the main thing you can take away from this is that the geometry minimisation has accurately give the main conformation of the various structures. Displaying the results in tables is fine, but for this kind of analysis a graphical representation can be beneficial (usually a bar chart) in the relevant literature. Since you calculated data for two different isomers it would have been interesting to see if you could differentiate them using these results – i.e. could you tell which unknown isomer you have given the NMR spectrum. The main reason for calculating the 13C NMR spectrum rather than the 1H NMR spectrum is that prediction of the 1H spectrum is more difficult. The last part of the MP was interesting because it is somewhat similar to the typical activities of computational chemists looking at reaction mechanisms: As in this paper, it is often important to consider different conformations of reactants to work out reaction mechanisms – although ultimately the key energy is probably that of the transition state because a higher energy conformation may be more reactive (consider elimination of HBr from bromocyclohexane – this can only occur if Br is axial). Overall, you have amassed an impressive amount of results but it may have been better to focus more on data analysis than acquisition. What you have definitely shown is that this set of compounds can be very accurately modelled.

Talk:Mod:SAMROWE001

2011-12-02T16:06:39Z

Mjhughes:

====Q1====
*Correct calculation for the dimers, but we're looking for some analysis as well. Where do you think the difference in energy come from? Which part of the molecules is responsible for it? The two hydrogens you mentioned is part of it, but there's more. You'll need to go back to the meaning of each component of the overall energy.
*Again, which part of the monohydrogenated compounds is responsible for the difference in bend energy?
====Q2====
*I appreciate that you discussed the twisted-boat conformers and explained how you arrived at the final optimised structures.
A bit more analysis on the MMFF94 data and components would be great here. Does it give the same structure as MM2?
*I think you 'get' the concept of hyperstable alkenes, but will need to explain it more clearly. Why didn't you show more calculation results of the hydrogenated compound? Can you compare that with the alkene (BIG CLUE: balance the equation!).
====Q3====
*Your picture size is good, but from this point of view I can't see the other half of the molecule. This is important when looking at MOs, so please adjust the pictures.
*A bit more elaboration on the σ*-π interaction (specifying each orbitals are involved and the consequences which you will observe later with your vibrational frequencies calculation) would be great.

====Q4====
*Which complex interactions are we talking about? and why can't MM2 handle them?
*From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?
*Your MM2 structures (which I can't find) aren't the lowest in energy, although your PM6 energies are close (not quite yet). However, it's the analysis and explanation that matter here.
*It would be very important to have a figure showing which structures are A/A' B/B' C/C' and D/D'. I already think you're using a different system from the one in the question.
*" it is able to detect that ring B' shows many unfavourable orbital interactions and therefore distorts the entire geometry to produce a structure which is very similar to that of ring B. " I don't think there is unfavourable interaction here. B is just so more stable. Did you start off with structure B from MM2? Again, since I'm not seeing the jmol, it's all guess work.
*The whole purpose of this question is to let students find out the origin of the diastereoselectivity observed for this kind of reaction. From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?

====Overall====
I think you made a good start, but the analysis side is still lacking for really good mark. Nevertheless, I hope the feeback help and would welcome any further discussion.

====After Marking:====

Q1: Your energy values are all correct and the analysis of the contributions to strain is broadly correct. For compounds 3 and 4 the main difference is in the alkene bond angles rather than the bond angles specifically within the bridged bicyclic part. What you say about kinetic and thermodynamic control is correct – you can’t really say much more about the hydrogenation than you have without more information about the reaction conditions and knowledge of the transition state energies (which would require DFT methods).

Q2. Your energy values and the structures of your global minima are correct. Although you have the right answer it would have been good to see how you got there – e.g. showing some higher energy conformations you found along the way. For example, although 6-rings often adopt chair conformations this is not always the most stable form (sometimes twist-boat or boat can be formed). Bending strain specifically refers to deviation from ideal bond angles so there will be some angles in isomer 9 that deviate more than in isomer 10. The definition of hyperstable alkenes is good and the qualitative analysis of strain in the hydrogenated form is a nice extra – it is not really possible to compare energy values using MM methods because these are not isomeric species (also you need to include the energy of hydrogen to get a balanced equation).

Q4. R=Me is the correct choice and the energy values you have using PM6 are pretty good; Some of the MM2 results are a bit off (there should still be a significant difference between A/B and A’/B’ (>10kcal/mol) ; without the structures it is hard to see why your energies are different – there is quite a degree of flexibility even in this relatively small system. As you have found, when using MOPAC methods A=C etc. You’re explanation for this is correct – MOPAC can determine bonding interactions if the appropriate atoms are close, whereas MM cannot. NB: The Burgi-Dunitz angle is 107.

MP. Your calculated spectral data all seems to give a good match to the reported values – the main thing you can take away from this is that the geometry minimisation has accurately give the main conformation of the various structures. Displaying the results in tables is fine, but for this kind of analysis a graphical representation can be beneficial (usually a bar chart) in the relevant literature. Since you calculated data for two different isomers it would have been interesting to see if you could differentiate them using these results – i.e. could you tell which unknown isomer you have given the NMR spectrum. The main reason for calculating the 13C NMR spectrum rather than the 1H NMR spectrum is that prediction of the 1H spectrum is more difficult. The last part of the MP was interesting because it is somewhat similar to the typical activities of computational chemists looking at reaction mechanisms: As in this paper, it is often important to consider different conformations of reactants to work out reaction mechanisms – although ultimately the key energy is probably that of the transition state because a higher energy conformation may be more reactive (consider elimination of HBr from bromocyclohexane – this can only occur if Br is axial). Overall, you have amassed an impressive amount of results but it may have been better to focus more on data analysis than acquisition. What you have definitely shown is that this set of compounds can be very accurately modelled.

Talk:Mod:SAMROWE001

2011-12-02T16:04:37Z

Mjhughes:

====Q1====
*Correct calculation for the dimers, but we're looking for some analysis as well. Where do you think the difference in energy come from? Which part of the molecules is responsible for it? The two hydrogens you mentioned is part of it, but there's more. You'll need to go back to the meaning of each component of the overall energy.
*Again, which part of the monohydrogenated compounds is responsible for the difference in bend energy?
====Q2====
*I appreciate that you discussed the twisted-boat conformers and explained how you arrived at the final optimised structures.
A bit more analysis on the MMFF94 data and components would be great here. Does it give the same structure as MM2?
*I think you 'get' the concept of hyperstable alkenes, but will need to explain it more clearly. Why didn't you show more calculation results of the hydrogenated compound? Can you compare that with the alkene (BIG CLUE: balance the equation!).
====Q3====
*Your picture size is good, but from this point of view I can't see the other half of the molecule. This is important when looking at MOs, so please adjust the pictures.
*A bit more elaboration on the σ*-π interaction (specifying each orbitals are involved and the consequences which you will observe later with your vibrational frequencies calculation) would be great.

====Q4====
*Which complex interactions are we talking about? and why can't MM2 handle them?
*From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?
*Your MM2 structures (which I can't find) aren't the lowest in energy, although your PM6 energies are close (not quite yet). However, it's the analysis and explanation that matter here.
*It would be very important to have a figure showing which structures are A/A' B/B' C/C' and D/D'. I already think you're using a different system from the one in the question.
*" it is able to detect that ring B' shows many unfavourable orbital interactions and therefore distorts the entire geometry to produce a structure which is very similar to that of ring B. " I don't think there is unfavourable interaction here. B is just so more stable. Did you start off with structure B from MM2? Again, since I'm not seeing the jmol, it's all guess work.
*The whole purpose of this question is to let students find out the origin of the diastereoselectivity observed for this kind of reaction. From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?

====Overall====
I think you made a good start, but the analysis side is still lacking for really good mark. Nevertheless, I hope the feeback help and would welcome any further discussion.

====After Marking:====

Q1: Your energy values are all correct and the analysis of the contributions to strain is broadly correct. For compounds 3 and 4 the main difference is in the alkene bond angles rather than the bond angles specifically within the bridged bicyclic part. What you say about kinetic and thermodynamic control is correct – you can’t really say much more about the hydrogenation than you have without more information about the reaction conditions and knowledge of the transition state energies (which would require DFT methods).

Q2. Your energy values and the structures of your global minima are correct. Although you have the right answer it would have been good to see how you got there – e.g. showing some higher energy conformations you found along the way. For example, although 6-rings often adopt chair conformations this is not always the most stable form (sometimes twist-boat or boat can be formed). Bending strain specifically refers to deviation from ideal bond angles so there will be some angles in isomer 9 that deviate more than in isomer 10. The definition of hyperstable alkenes is good and the qualitative analysis of strain in the hydrogenated form is a nice extra – it is not really possible to compare energy values using MM methods because these are not isomeric species (also you need to include the energy of hydrogen to get a balanced equation).

Q3. The MOs look good and the rationalisation of reactivity is correct. The stretching frequencies are a little higher than expected for the dialkene. Without seeing them it is hard to say why this is the case. Possibly these values don’t correspond to the actual C-Cl and C=C stretches – the various vibrations can look quite similar from certain angles, but the stretching should show clear compression and relaxation of the bonds in question. The C-Cl bond is in fact expected to be stronger in the monoalkene; this is because the double bond that is removed is capable of interacting with the C-Cl sigma* orbital through its pi orbital and this interaction weakens the C-Cl bond. For any calculations you carry out with different methods, it is worth including the energy values you obtain – these can be used to assess the accuracy of the results.

Q4. R=Me is the correct choice and the energy values you have using PM6 are pretty good; Some of the MM2 results are a bit off (there should still be a significant difference between A/B and A’/B’ (>10kcal/mol) ; without the structures it is hard to see why your energies are different – there is quite a degree of flexibility even in this relatively small system. As you have found, when using MOPAC methods A=C etc. You’re explanation for this is correct – MOPAC can determine bonding interactions if the appropriate atoms are close, whereas MM cannot. NB: The Burgi-Dunitz angle is 107.

MP. Your calculated spectral data all seems to give a good match to the reported values – the main thing you can take away from this is that the geometry minimisation has accurately give the main conformation of the various structures. Displaying the results in tables is fine, but for this kind of analysis a graphical representation can be beneficial (usually a bar chart) in the relevant literature. Since you calculated data for two different isomers it would have been interesting to see if you could differentiate them using these results – i.e. could you tell which unknown isomer you have given the NMR spectrum. The main reason for calculating the 13C NMR spectrum rather than the 1H NMR spectrum is that prediction of the 1H spectrum is more difficult. The last part of the MP was interesting because it is somewhat similar to the typical activities of computational chemists looking at reaction mechanisms: As in this paper, it is often important to consider different conformations of reactants to work out reaction mechanisms – although ultimately the key energy is probably that of the transition state because a higher energy conformation may be more reactive (consider elimination of HBr from bromocyclohexane – this can only occur if Br is axial). Overall, you have amassed an impressive amount of results but it may have been better to focus more on data analysis than acquisition. What you have definitely shown is that this set of compounds can be very accurately modelled.

Talk:Mod:SAMROWE001

2011-12-02T16:03:46Z

Mjhughes:

====Q1====
*Correct calculation for the dimers, but we're looking for some analysis as well. Where do you think the difference in energy come from? Which part of the molecules is responsible for it? The two hydrogens you mentioned is part of it, but there's more. You'll need to go back to the meaning of each component of the overall energy.
*Again, which part of the monohydrogenated compounds is responsible for the difference in bend energy?
====Q2====
*I appreciate that you discussed the twisted-boat conformers and explained how you arrived at the final optimised structures.
A bit more analysis on the MMFF94 data and components would be great here. Does it give the same structure as MM2?
*I think you 'get' the concept of hyperstable alkenes, but will need to explain it more clearly. Why didn't you show more calculation results of the hydrogenated compound? Can you compare that with the alkene (BIG CLUE: balance the equation!).
====Q3====
*Your picture size is good, but from this point of view I can't see the other half of the molecule. This is important when looking at MOs, so please adjust the pictures.
*A bit more elaboration on the σ*-π interaction (specifying each orbitals are involved and the consequences which you will observe later with your vibrational frequencies calculation) would be great.

====Q4====
*Which complex interactions are we talking about? and why can't MM2 handle them?
*From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?
*Your MM2 structures (which I can't find) aren't the lowest in energy, although your PM6 energies are close (not quite yet). However, it's the analysis and explanation that matter here.
*It would be very important to have a figure showing which structures are A/A' B/B' C/C' and D/D'. I already think you're using a different system from the one in the question.
*" it is able to detect that ring B' shows many unfavourable orbital interactions and therefore distorts the entire geometry to produce a structure which is very similar to that of ring B. " I don't think there is unfavourable interaction here. B is just so more stable. Did you start off with structure B from MM2? Again, since I'm not seeing the jmol, it's all guess work.
*The whole purpose of this question is to let students find out the origin of the diastereoselectivity observed for this kind of reaction. From the energies for intermediate C/C' and D/D' can you calculate their equilibrium constants and predict the diastereoselectivity of the reactions?

====Overall====
I think you made a good start, but the analysis side is still lacking for really good mark. Nevertheless, I hope the feeback help and would welcome any further discussion.

After Marking:

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your energy values are all correct and the analysis of the contributions to strain is broadly correct. For compounds 3 and 4 the main difference is in the alkene bond angles rather than the bond angles specifically within the bridged bicyclic part. What you say about kinetic and thermodynamic control is correct – you can’t really say much more about the hydrogenation than you have without more information about the reaction conditions and knowledge of the transition state energies (which would require DFT methods).

Q2. Your energy values and the structures of your global minima are correct. Although you have the right answer it would have been good to see how you got there – e.g. showing some higher energy conformations you found along the way. For example, although 6-rings often adopt chair conformations this is not always the most stable form (sometimes twist-boat or boat can be formed). Bending strain specifically refers to deviation from ideal bond angles so there will be some angles in isomer 9 that deviate more than in isomer 10. The definition of hyperstable alkenes is good and the qualitative analysis of strain in the hydrogenated form is a nice extra – it is not really possible to compare energy values using MM methods because these are not isomeric species (also you need to include the energy of hydrogen to get a balanced equation).

Q3. The MOs look good and the rationalisation of reactivity is correct. The stretching frequencies are a little higher than expected for the dialkene. Without seeing them it is hard to say why this is the case. Possibly these values don’t correspond to the actual C-Cl and C=C stretches – the various vibrations can look quite similar from certain angles, but the stretching should show clear compression and relaxation of the bonds in question. The C-Cl bond is in fact expected to be stronger in the monoalkene; this is because the double bond that is removed is capable of interacting with the C-Cl sigma* orbital through its pi orbital and this interaction weakens the C-Cl bond. For any calculations you carry out with different methods, it is worth including the energy values you obtain – these can be used to assess the accuracy of the results.

Q4. R=Me is the correct choice and the energy values you have using PM6 are pretty good; Some of the MM2 results are a bit off (there should still be a significant difference between A/B and A’/B’ (>10kcal/mol) ; without the structures it is hard to see why your energies are different – there is quite a degree of flexibility even in this relatively small system. As you have found, when using MOPAC methods A=C etc. You’re explanation for this is correct – MOPAC can determine bonding interactions if the appropriate atoms are close, whereas MM cannot. NB: The Burgi-Dunitz angle is 107.

MP. Your calculated spectral data all seems to give a good match to the reported values – the main thing you can take away from this is that the geometry minimisation has accurately give the main conformation of the various structures. Displaying the results in tables is fine, but for this kind of analysis a graphical representation can be beneficial (usually a bar chart) in the relevant literature. Since you calculated data for two different isomers it would have been interesting to see if you could differentiate them using these results – i.e. could you tell which unknown isomer you have given the NMR spectrum. The main reason for calculating the 13C NMR spectrum rather than the 1H NMR spectrum is that prediction of the 1H spectrum is more difficult. The last part of the MP was interesting because it is somewhat similar to the typical activities of computational chemists looking at reaction mechanisms: As in this paper, it is often important to consider different conformations of reactants to work out reaction mechanisms – although ultimately the key energy is probably that of the transition state because a higher energy conformation may be more reactive (consider elimination of HBr from bromocyclohexane – this can only occur if Br is axial). Overall, you have amassed an impressive amount of results but it may have been better to focus more on data analysis than acquisition. What you have definitely shown is that this set of compounds can be very accurately modelled.

Talk:Mod:cherrybakewell

2011-12-02T16:02:28Z

Mjhughes:

Q1: All of your energy values are correct and the analysis of strain contribution is good. The discussion of kinetic vs thermodynamic control is all correct – as you say you can’t say much about the outcome of the hydrogenation without further mechanistic information and transition state calculations.

Q2. The structures and energies you have are correct but it would have been nice to see some discussion about how you achieved this minimisation – e.g. showing other conformations found and the key features that needed tweaking. You correctly identified the 6-ring conformation as a key area to look at - the twist boat forms would have been good to look at because the chair is usually the favoured conformation for substituted cyclohexanes but not always. The analysis of the stability of these compounds is correct – the parent alkane experiences a lot of strain due to the type of ring system; it is specifically a “hyperstable olefin”. NB: the reaction you are talking about here is a hydrogenation not a hydration which would be addition of water over the double bond (e.g. one -OH group one -H group).

Q3.Your MOs and IR stretches are correct. The more reactive alkene (towards electrophiles) is correctly identified. The C-Cl bond is indeed weakened by the pi-simga* interaction that you describe. One thing worth noting is that you should include all of the energies that you calculate (e.g. MOPAC as well as MM) because these are used as a judge of the calculations to some extent.

Q4. Your energy values are excellent (only one of the MM2 calculations is a little off (C)) and I can’t really see what’s wrong with the structure. As you said with MOPAC A=C and B=D – this is because the bonding interaction of proximal groups can be determined (unlike with MM2 in which the bonds are fixed at the start). Having obtained some energy values and structures for C/C’/D/D’ you could have commented further on the stereoselectivity of the glycosidation. C, C’, D and D’ all favour one product after nucleophilic attack and the distribution will be heavily in favour of C and D since they are much lower in energy to C’ and D’; additionally the trajectory of attack is better in these lower energy configurations.

MP. Your discussion of the geometry optimisation is good – one aim from the course is to use the methods from the set questions in the more open ended questions. For your NMR data: presenting the information in tables is fine and it is good that you have given the deviations from lit values, but there should be some more discussion in the text that describes the accuracy of the results. There are also more creative ways of performing this analysis - this sort of data is usually depicted graphically (bar charts are typical). Since you have calculated data for both isomers it would also have been possible to determine whether the calculation can distinguish between them. The most useful application of this type of calculation is being able to work out which isomer you have given the NMR of an unknown compound. With the IR data – comparison with the lit is necessary to tell if it accurate (admittedly the IR data is not usually extensive in synthetic reports).

Talk:Mod:cherrybakewell

2011-12-02T16:02:11Z

Mjhughes:

Q1: All of your energy values are correct and the analysis of strain contribution is good. The discussion of kinetic vs thermodynamic control is all correct – as you say you can’t say much about the outcome of the hydrogenation without further mechanistic information and transition state calculations.

Q2. The structures and energies you have are correct but it would have been nice to see some discussion about how you achieved this minimisation – e.g. showing other conformations found and the key features that needed tweaking. You correctly identified the 6-ring conformation as a key area to look at - the twist boat forms would have been good to look at because the chair is usually the favoured conformation for substituted cyclohexanes but not always. The analysis of the stability of these compounds is correct – the parent alkane experiences a lot of strain due to the type of ring system; it is specifically a “hyperstable olefin”. NB: the reaction you are talking about here is a hydrogenation not a hydration which would be addition of water over the double bond (e.g. one -OH group one -H group).

Q3. Normal 0 false false false EN-GB X-NONE X-NONE

Your MOs and IR stretches are correct. The more reactive alkene (towards electrophiles) is correctly identified. The C-Cl bond is indeed weakened by the pi-simga* interaction that you describe. One thing worth noting is that you should include all of the energies that you calculate (e.g. MOPAC as well as MM) because these are used as a judge of the calculations to some extent.

Q4. Your energy values are excellent (only one of the MM2 calculations is a little off (C)) and I can’t really see what’s wrong with the structure. As you said with MOPAC A=C and B=D – this is because the bonding interaction of proximal groups can be determined (unlike with MM2 in which the bonds are fixed at the start). Having obtained some energy values and structures for C/C’/D/D’ you could have commented further on the stereoselectivity of the glycosidation. C, C’, D and D’ all favour one product after nucleophilic attack and the distribution will be heavily in favour of C and D since they are much lower in energy to C’ and D’; additionally the trajectory of attack is better in these lower energy configurations.

MP. Your discussion of the geometry optimisation is good – one aim from the course is to use the methods from the set questions in the more open ended questions. For your NMR data: presenting the information in tables is fine and it is good that you have given the deviations from lit values, but there should be some more discussion in the text that describes the accuracy of the results. There are also more creative ways of performing this analysis - this sort of data is usually depicted graphically (bar charts are typical). Since you have calculated data for both isomers it would also have been possible to determine whether the calculation can distinguish between them. The most useful application of this type of calculation is being able to work out which isomer you have given the NMR of an unknown compound. With the IR data – comparison with the lit is necessary to tell if it accurate (admittedly the IR data is not usually extensive in synthetic reports).

Talk:Mod:Pterodactyladiene

2011-12-02T16:00:57Z

Mjhughes:

Q1: Your energy values are spot on and the discussion of strain contributions is good. It is not necessary to give so many decimal places in the results because these are approximations with some degree of error. You are right that the discussion of thermodynamic vs kinetic control can only be limited: The dimerization must be under kinetic control because the lowest energy molecule is not formed. The hydrogenation will give molecule 4 if the reaction is under thermodynamic control, but under other conditions information about the mechanism and transition state calculations will be needed.

Q2. While atropisomerism is often exhibited in cases where there is restricted rotation of aromatic rings (e.g. BINOL), this is not the case here – there are no aromatic rings! It is just a case of restricted rotation in a different sense (it is difficult for the carbonyl group to flip its orientation because the barrier to rotation is high). Your energies and structures are good, but it would have been nice to see some discussion about how you came to these results – e.g. showing some higher energy conformations that you found and looking at what was the key structural feature to change. The definition of a hyperstable alkene is correct – calculation of the olefin strain is a good extra although it would be better to use DFT methods to compare the relative energies

Q3. The MOs all look good – I wouldn’t have chosen that orientation to show them however (it’s a bit of a nit picking point but it would be better to see them at an angle to see some perspective). The reactivity is correctly described with reference to the HOMO. The IR stretches look fine and the effect of the pi-sigma* interaction is well analysed.

Q4. It seems that you stopped halfway through your answer at the end; perhaps this was due to time constraint, but otherwise make sure to check over what you have done before submission. R=Me is the correct choice here and your energy values are pretty good (especially the PM6 ones); for MM2 some values are a bit high, but it is more difficult to find the global minimum (as you will have seen in Q2). Some of your jmols don’t match the description – be careful to get the correct file name when putting in the code. Your calculations show (in terms of both energies and structures) that A=C and B=D when MOPAC calculations are performed – this is because when the carbonyl oxygen is close to the oxonium carbon, bonding interactions can be determined. In contrast in MM2 the bonds are fixed by the user at the beginning.

MP. The discussion of the energetics of the different isomers is good. This application of methods used in the earlier questions is what we hope to see in the more open ended questions. The NMR and IR data looks to be a reasonable match. It would be worth considering if the inaccuracies are due to conformational flexibility by seeing if there are any other low energy conformations which could contribute to the physical properties of the compounds. More deshielded nuclei would be expected to be further downfield (i.e. higher ppm); the reason that you’re other peaks appear to be more deshielded than the one at 52 is that they are all in aromatic systems. Since you calculated NMR spectra for both isomers it would have been nice to see if the calculation could be used to distinguish one from the other (given a spectrum of an unknown). This would involve comparison of the two sets of computational spectra against the experimental spectrum and seeing if one example gives a better fit. For this kind of error analysis it can be useful if it’s done graphically (bar charts are often used) – presentation of the data in tables is fine, but different approaches can be considered (perhaps by consulting reported computation NMR data).

Talk:Mod:cherrybakewell

2011-12-02T16:00:00Z

Mjhughes:

Q1: All of your energy values are correct and the analysis of strain contribution is good. The discussion of kinetic vs thermodynamic control is all correct – as you say you can’t say much about the outcome of the hydrogenation without further mechanistic information and transition state calculations.

Q2. The structures and energies you have are correct but it would have been nice to see some discussion about how you achieved this minimisation – e.g. showing other conformations found and the key features that needed tweaking. You correctly identified the 6-ring conformation as a key area to look at - the twist boat forms would have been good to look at because the chair is usually the favoured conformation for substituted cyclohexanes but not always. The analysis of the stability of these compounds is correct – the parent alkane experiences a lot of strain due to the type of ring system; it is specifically a “hyperstable olefin”. NB: the reaction you are talking about here is a hydrogenation not a hydration which would be addition of water over the double bond (e.g. one -OH group one -H group).

Q4. Your energy values are excellent (only one of the MM2 calculations is a little off (C)) and I can’t really see what’s wrong with the structure. As you said with MOPAC A=C and B=D – this is because the bonding interaction of proximal groups can be determined (unlike with MM2 in which the bonds are fixed at the start). Having obtained some energy values and structures for C/C’/D/D’ you could have commented further on the stereoselectivity of the glycosidation. C, C’, D and D’ all favour one product after nucleophilic attack and the distribution will be heavily in favour of C and D since they are much lower in energy to C’ and D’; additionally the trajectory of attack is better in these lower energy configurations.

MP. Your discussion of the geometry optimisation is good – one aim from the course is to use the methods from the set questions in the more open ended questions. For your NMR data: presenting the information in tables is fine and it is good that you have given the deviations from lit values, but there should be some more discussion in the text that describes the accuracy of the results. There are also more creative ways of performing this analysis - this sort of data is usually depicted graphically (bar charts are typical). Since you have calculated data for both isomers it would also have been possible to determine whether the calculation can distinguish between them. The most useful application of this type of calculation is being able to work out which isomer you have given the NMR of an unknown compound. With the IR data – comparison with the lit is necessary to tell if it accurate (admittedly the IR data is not usually extensive in synthetic reports).

Talk:Mod:Pterodactyladiene

2011-12-02T15:57:26Z

Mjhughes:

Q1: Your energy values are spot on and the discussion of strain contributions is good. It is not necessary to give so many decimal places in the results because these are approximations with some degree of error. You are right that the discussion of thermodynamic vs kinetic control can only be limited: The dimerization must be under kinetic control because the lowest energy molecule is not formed. The hydrogenation will give molecule 4 if the reaction is under thermodynamic control, but under other conditions information about the mechanism and transition state calculations will be needed.

Q2. While atropisomerism is often exhibited in cases where there is restricted rotation of aromatic rings (e.g. BINOL), this is not the case here – there are no aromatic rings! It is just a case of restricted rotation in a different sense (it is difficult for the carbonyl group to flip its orientation because the barrier to rotation is high). Your energies and structures are good, but it would have been nice to see some discussion about how you came to these results – e.g. showing some higher energy conformations that you found and looking at what was the key structural feature to change. The definition of a hyperstable alkene is correct – calculation of the olefin strain is a good extra although it would be better to use DFT methods to compare the relative energies

Q3. Your MOs and IR stretches are correct. The more reactive alkene (towards electrophiles) is correctly identified. The C-Cl bond is indeed weakened by the pi-simga* interaction that you describe. One thing worth noting is that you should include all of the energies that you calculate (e.g. MOPAC as well as MM) because these are used as a judge of the calculations to some extent.

Q4. It seems that you stopped halfway through your answer at the end; perhaps this was due to time constraint, but otherwise make sure to check over what you have done before submission. R=Me is the correct choice here and your energy values are pretty good (especially the PM6 ones); for MM2 some values are a bit high, but it is more difficult to find the global minimum (as you will have seen in Q2). Some of your jmols don’t match the description – be careful to get the correct file name when putting in the code. Your calculations show (in terms of both energies and structures) that A=C and B=D when MOPAC calculations are performed – this is because when the carbonyl oxygen is close to the oxonium carbon, bonding interactions can be determined. In contrast in MM2 the bonds are fixed by the user at the beginning.

MP. The discussion of the energetics of the different isomers is good. This application of methods used in the earlier questions is what we hope to see in the more open ended questions. The NMR and IR data looks to be a reasonable match. It would be worth considering if the inaccuracies are due to conformational flexibility by seeing if there are any other low energy conformations which could contribute to the physical properties of the compounds. More deshielded nuclei would be expected to be further downfield (i.e. higher ppm); the reason that you’re other peaks appear to be more deshielded than the one at 52 is that they are all in aromatic systems. Since you calculated NMR spectra for both isomers it would have been nice to see if the calculation could be used to distinguish one from the other (given a spectrum of an unknown). This would involve comparison of the two sets of computational spectra against the experimental spectrum and seeing if one example gives a better fit. For this kind of error analysis it can be useful if it’s done graphically (bar charts are often used) – presentation of the data in tables is fine, but different approaches can be considered (perhaps by consulting reported computation NMR data).

Talk:Mod:Pterodactyladiene

2011-12-02T15:57:09Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your energy values are spot on and the discussion of strain contributions is good. It is not necess..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your energy values are spot on and the discussion of strain contributions is good. It is not necessary to give so many decimal places in the results because these are approximations with some degree of error. You are right that the discussion of thermodynamic vs kinetic control can only be limited: The dimerization must be under kinetic control because the lowest energy molecule is not formed. The hydrogenation will give molecule 4 if the reaction is under thermodynamic control, but under other conditions information about the mechanism and transition state calculations will be needed.

Q2. While atropisomerism is often exhibited in cases where there is restricted rotation of aromatic rings (e.g. BINOL), this is not the case here – there are no aromatic rings! It is just a case of restricted rotation in a different sense (it is difficult for the carbonyl group to flip its orientation because the barrier to rotation is high). Your energies and structures are good, but it would have been nice to see some discussion about how you came to these results – e.g. showing some higher energy conformations that you found and looking at what was the key structural feature to change. The definition of a hyperstable alkene is correct – calculation of the olefin strain is a good extra although it would be better to use DFT methods to compare the relative energies

Q3. Your MOs and IR stretches are correct. The more reactive alkene (towards electrophiles) is correctly identified. The C-Cl bond is indeed weakened by the pi-simga* interaction that you describe. One thing worth noting is that you should include all of the energies that you calculate (e.g. MOPAC as well as MM) because these are used as a judge of the calculations to some extent.

Q4. It seems that you stopped halfway through your answer at the end; perhaps this was due to time constraint, but otherwise make sure to check over what you have done before submission. R=Me is the correct choice here and your energy values are pretty good (especially the PM6 ones); for MM2 some values are a bit high, but it is more difficult to find the global minimum (as you will have seen in Q2). Some of your jmols don’t match the description – be careful to get the correct file name when putting in the code. Your calculations show (in terms of both energies and structures) that A=C and B=D when MOPAC calculations are performed – this is because when the carbonyl oxygen is close to the oxonium carbon, bonding interactions can be determined. In contrast in MM2 the bonds are fixed by the user at the beginning.

MP. The discussion of the energetics of the different isomers is good. This application of methods used in the earlier questions is what we hope to see in the more open ended questions. The NMR and IR data looks to be a reasonable match. It would be worth considering if the inaccuracies are due to conformational flexibility by seeing if there are any other low energy conformations which could contribute to the physical properties of the compounds. More deshielded nuclei would be expected to be further downfield (i.e. higher ppm); the reason that you’re other peaks appear to be more deshielded than the one at 52 is that they are all in aromatic systems. Since you calculated NMR spectra for both isomers it would have been nice to see if the calculation could be used to distinguish one from the other (given a spectrum of an unknown). This would involve comparison of the two sets of computational spectra against the experimental spectrum and seeing if one example gives a better fit. For this kind of error analysis it can be useful if it’s done graphically (bar charts are often used) – presentation of the data in tables is fine, but different approaches can be considered (perhaps by consulting reported computation NMR data).

Talk:Mod:cherrybakewell

2011-12-02T15:56:38Z

Mjhughes:

Q1: All of your energy values are correct and the analysis of strain contribution is good. The discussion of kinetic vs thermodynamic control is all correct – as you say you can’t say much about the outcome of the hydrogenation without further mechanistic information and transition state calculations.

Q2. The structures and energies you have are correct but it would have been nice to see some discussion about how you achieved this minimisation – e.g. showing other conformations found and the key features that needed tweaking. You correctly identified the 6-ring conformation as a key area to look at - the twist boat forms would have been good to look at because the chair is usually the favoured conformation for substituted cyclohexanes but not always. The analysis of the stability of these compounds is correct – the parent alkane experiences a lot of strain due to the type of ring system; it is specifically a “hyperstable olefin”. NB: the reaction you are talking about here is a hydrogenation not a hydration which would be addition of water over the double bond (e.g. one -OH group one -H group).

Q3. The MOs all look good – I wouldn’t have chosen that orientation to show them however (it’s a bit of a nit picking point but it would be better to see them at an angle to see some perspective). The reactivity is correctly described with reference to the HOMO. The IR stretches look fine and the effect of the pi-sigma* interaction is well analysed.

Q4. Your energy values are excellent (only one of the MM2 calculations is a little off (C)) and I can’t really see what’s wrong with the structure. As you said with MOPAC A=C and B=D – this is because the bonding interaction of proximal groups can be determined (unlike with MM2 in which the bonds are fixed at the start). Having obtained some energy values and structures for C/C’/D/D’ you could have commented further on the stereoselectivity of the glycosidation. C, C’, D and D’ all favour one product after nucleophilic attack and the distribution will be heavily in favour of C and D since they are much lower in energy to C’ and D’; additionally the trajectory of attack is better in these lower energy configurations.

MP. Your discussion of the geometry optimisation is good – one aim from the course is to use the methods from the set questions in the more open ended questions. For your NMR data: presenting the information in tables is fine and it is good that you have given the deviations from lit values, but there should be some more discussion in the text that describes the accuracy of the results. There are also more creative ways of performing this analysis - this sort of data is usually depicted graphically (bar charts are typical). Since you have calculated data for both isomers it would also have been possible to determine whether the calculation can distinguish between them. The most useful application of this type of calculation is being able to work out which isomer you have given the NMR of an unknown compound. With the IR data – comparison with the lit is necessary to tell if it accurate (admittedly the IR data is not usually extensive in synthetic reports).

Talk:Mod:cherrybakewell

2011-12-02T15:56:11Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: All of your energy values are correct and the analysis of strain contribution is good. The discussi..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: All of your energy values are correct and the analysis of strain contribution is good. The discussion of kinetic vs thermodynamic control is all correct – as you say you can’t say much about the outcome of the hydrogenation without further mechanistic information and transition state calculations.
Q2. The structures and energies you have are correct but it would have been nice to see some discussion about how you achieved this minimisation – e.g. showing other conformations found and the key features that needed tweaking. You correctly identified the 6-ring conformation as a key area to look at - the twist boat forms would have been good to look at because the chair is usually the favoured conformation for substituted cyclohexanes but not always. The analysis of the stability of these compounds is correct – the parent alkane experiences a lot of strain due to the type of ring system; it is specifically a “hyperstable olefin”. NB: the reaction you are talking about here is a hydrogenation not a hydration which would be addition of water over the double bond (e.g. one -OH group one -H group).
Q3. The MOs all look good – I wouldn’t have chosen that orientation to show them however (it’s a bit of a nit picking point but it would be better to see them at an angle to see some perspective). The reactivity is correctly described with reference to the HOMO. The IR stretches look fine and the effect of the pi-sigma* interaction is well analysed.
Q4. Your energy values are excellent (only one of the MM2 calculations is a little off (C)) and I can’t really see what’s wrong with the structure. As you said with MOPAC A=C and B=D – this is because the bonding interaction of proximal groups can be determined (unlike with MM2 in which the bonds are fixed at the start). Having obtained some energy values and structures for C/C’/D/D’ you could have commented further on the stereoselectivity of the glycosidation. C, C’, D and D’ all favour one product after nucleophilic attack and the distribution will be heavily in favour of C and D since they are much lower in energy to C’ and D’; additionally the trajectory of attack is better in these lower energy configurations.
MP. Your discussion of the geometry optimisation is good – one aim from the course is to use the methods from the set questions in the more open ended questions. For your NMR data: presenting the information in tables is fine and it is good that you have given the deviations from lit values, but there should be some more discussion in the text that describes the accuracy of the results. There are also more creative ways of performing this analysis - this sort of data is usually depicted graphically (bar charts are typical). Since you have calculated data for both isomers it would also have been possible to determine whether the calculation can distinguish between them. The most useful application of this type of calculation is being able to work out which isomer you have given the NMR of an unknown compound. With the IR data – comparison with the lit is necessary to tell if it accurate (admittedly the IR data is not usually extensive in synthetic reports).

Talk:Mod1: Lukas Miseikis

2011-12-02T15:54:29Z

Mjhughes: Created page with "Q1: Your energy values are all correct and your analysis of strain contributions is good. For the hydrogenation products the major difference is the bending strain due to differ..."

Q1: Your energy values are all correct and your analysis of strain contributions is good. For the hydrogenation products the major difference is the bending strain due to different alkene bond angles which deviate from ideality to a different extent. You state “in this case Endo is the thermodynamic product” when I think you mean the opposite on the basis of the rest of the paragraph and your results – be careful that the key statements are worded correctly! What you say about kinetic vs thermodynamic control is good. The requirement for kinetic control is not really that certain products are inaccessible, it is more that the reaction is irreversible (under the time period of the reaction equilibrium between the products and starting materials is not reached); i.e. reactions can be poorly selective but still under kinetic control.

Q2. Your global minimum energy values and structures are correct and the general approach to answering this question is good (detailing the way you came to the lowest energy conformation); the 6-ring is an obvious starting point because of the known conformations it can adopt. These compounds are indeed examples of hyperstable olefins and your calculation of OS is a nice extra (although it would have been better to use DFT methods to ensure an accurate comparison). Olefin strain is mostly an effect of unusual strain exhibited in the hydrogenated compounds (which would be reflected in both the product and transition state energies). The alkenes in this type of ring system are actually surprisingly flat and close to ideal dimensions in general.

Q3. The MOs look good and the IR stretches are correct. All of the explanations are reasonable: The C-Cl is indeed weakened when there is an exo double bond because of the pi-sigma* interaction and the endo alkene is the most reactive towards electrophiles. One little point – it is worth including all energy values calculated (whether for MM or MOPAC methods these are sometimes used as a gauge of the calculation accuracy during marking).

Q4. R=methyl is the correct choice here and your energy values are good. Some of the MM2 structures have lower energy conformations available, but they are harder to find (cf Q2). Your discussion about the differences between MM2 and MOPAC is correct – MOPAC can find bonding interactions even if you don’t explicitly draw them (hence A=C and B=D). Having calculated energies of C, C’, D and D’ you could have commented further on the reaction selectivity. The relative abundance of these key intermediates is an important factor in determining the reaction outcome as they each favour one product after nucleophilic attack. As well as this, the C and D configurations are more reactive because the trajectory for nucleophilic attack is better.

MP. This looks like a good choice for a mini-project and the question you are tackling is a good one – whether it is possible to distinguish between different isomers using calculated NMR. Presenting the data in tables is fine, but there are some better, graphical ways to do it (typically bar charts are used). You have stated that the lit data more closely matches the calculated data for the correct isomer and this type of analysis can help to present the case because it is easier to assess the deviations visually than by looking at a set of numbers. For the coupling constant, the ability to calculate it is only really useful if you can show that one isomer is substantially different from the other. One other aspect that would have been interesting to discuss is the way in which the authors of the report differentiated the isomers experimentally (usually 2D NMR techniques). A summary at the end of the mini project would have been helpful – also it may have been useful to you to set out your aims before answering the question.

Talk:Mod:khlmod1

2011-12-02T15:45:45Z

Mjhughes: Created page with "Q1: Your energy values are correct and the discussion about different strain contributions is good. You should be careful that the key phrases are correctly worded: you state ..."

Q1: Your energy values are correct and the discussion about different strain contributions is good. You should be careful that the key phrases are correctly worded: you state “Therefore, product 3 was found to have the lowest overall energy” when this is clearly the opposite of what you mean. What you say about kinetic vs thermodynamic control is right – indeed there isn’t much you can say about the outcome of the hydrogenation without further information about conditions, mechanism and with some transition state calculations (requiring DFT methods).

Q2. There are some problems with your structures – none of them actually represent the correct configuration of the intermediate. The aspects that are wrong in all or some of the structures are: double bond geometry, cis vs trans ring fusion (supposed to be cis i.e. hydrogens on same side) and relationship between ring fusion hydrogens and the dimethyl bridging group (should be on the same side). When performing these kinds of minimisations it is always important to keep an eye on the structure as it can easily become distorted away from what was intended. That said your approach to this question was exactly what was intended – analysis of different conformations to get a rationalised global minimisation and comparison of the strain contributions as in Q1. The extra material on the hyperstable olefin aspect is good - although accurate energy comparison is best achieved with DFT method, the qualitative consideration of the strain in the parent hydrocarbon is a good approach.

Q3. The MOs look good. The alkene syn to the chlorine is indeed the most reactive towards electrophiles. I think this is what you mean but the wording is a little confused; “HOMO is very susceptible to electrophilic attack” and “LUMO is very susceptible to nucleophilic attack” are tautologies – the important fact is which alkene does what (on the basis of its orbital distribution in the HOMO and LUMO). The IR stretches are good and the differences are well described and explained.

Q4. R=Me is indeed the best choice here to minimise job times. Your PM6 energies and structures are all good, although the MM2 ones are a little higher than expected – it is difficult to comment further on this without seeing the jmols, but it is more difficult to find the lowest energy conformer using MM2 in this task. Although you described the origin of the stereoselectivity at the beginning you didn’t correlate your findings back to this. You have found that there is an overwhelming preference for the cis fused intermediates and this means that nearly all of the reactants go through these intermediates and give the corresponding selectivity. Looking at the structures of the different intermediates, these lower energy conformations also have the best trajectory for attack so they are also more reactive. What you may have noticed is that using PM6 A=C and B=D; the method is capable of recognising the bonding interaction when the acetyl oxygen is close to the oxonium carbon (MM is not capable of this and the bonds are set from the start).

MP. The discussion of geometry minimisation is good – it is hoped that methods used in set questions will be applied to the more open-ended questions as much as possible. Your NMR data is a reasonably good match. Although it is fine to tabulate this data, it can be better presented in graphical form (typically bar charts are used). It is a shame that you did not attempt to compare the calculated data for both isomers with each set of lit data. In this way you could tell to what extent you can differentiate the two isomers and if you could work out the structure from a spectrum of an unknown isomer. With the IR data you need to compare to some experimental data to see if the calculation is accurate (it could be that the difference you see between the isomers is smaller than the error). A note on nomenclature: -Ph = phenyl or benzene ring (not benzyl ring – which is confusing given that a benzyl group is also present).

Mod:organic

2011-11-18T12:51:41Z

Mjhughes: /* : Structure based Mini project using DFT-based Molecular orbital methods */

See also: [[Mod:timetable|Timetable]],[[Mod:lectures|Intro lecture]],[[mod:laptop|Laptop use]], [[mod:programs|Programs]], [[mod:organic|Module 1]], [[Mod:inorganic|Module 2]], [[Mod:phys3|Module 3]],[[Mod:writeup|Writing up]], [[Mod:dont_panic|Don't panic]].

----
This module comes in two parts, examples of the use of the basic tools, followed by a mini-project.
= The basic techniques of molecular mechanics and semi-empirical molecular orbital methods for structural and spectroscopic evaluations =

==Objectives of this module of the course:==
It is now possible using a computer to accurately model many aspects of organic structure and reactivity, and such modelling can often be used not only to rationalise the outcomes of reactions, but to predict useful modifications or even new types of reaction. The selection of (short) modelling experiments contained in this module of the course attempts to illustrate some of the diversity of such molecular modelling. The module ends with a more open-ended exploration or ''mini-research'' exercise typical of that very often experienced in synthetic chemistry labs; namely is the structure of my final product correct?
#To use molecular mechanics (< 10 hours) to predict the geometry and regioselectivity of:
##the hydrogenation of cyclopentadiene dimer
##the stereochemistry of nucleophilic addition to two different pyridinium rings([[organic:NAD|NAD+ analogues]])
##the conformation/atropisomerism of a large ring ketone intermediate in one synthesis of the anti-cancer drug Taxol
#To use semi-empirical and DFT molecular orbital theory (< 10 hours) to investigate:
##the origins of the regioselectivity of the electrophilic carbenylation of a chloro-substituted bicyclic diene,
##the use of DFT molecular orbital theory to investigate Neighbouring group participation ([[organic:NGP|NGP]]) on the C-Cl and/or C=C stretching frequency of the above bicyclic diene
##concluding with a Mini-project investigating spectroscopic simulation in an organic molecule.
#To gain familiarity with the use of a institutional digital repository (< 5 minutes)
#To perform searches of the literature for each topic in order to cite in your final report any relevant references to each experiment as appropriate (< 1 hour)
#To present the results in the form of a Wiki page, with credit given for any annotation of these course notes, either to pages created here, or Wikipedia entries.
=== Background information ===
A general coverage of many of the topics in this module can be found in the [http://www.ch.ic.ac.uk/local/organic/mod 3rd year course on molecular modelling]. Podcasts (including slides and additional links to other related materials) can be found at this site.

A [http://www.mendeley.com/groups/4657/molecular-modelling-course/ Mendeley shared course page] has been set up as a pilot project. You will have to register with Mendeley to participate.

==Modelling using Molecular Mechanics==

A general introduction to the [[mod:molecular_mechanics|Molecular Mechanics]] (MM) method should be consulted before attempting any calculations. The present techniques illustrate several more complex applications of this method to typical chemical problems and the type of information that such modelling is capable of providing. This involves optimising molecular geometry to an energy minimum and analysing the final energy in terms of bond length and angle strain, steric effects and van der Waals contributions.

Before discussing specific applications of such a model, it is worth noting some of the limitations of the molecular mechanics approach. It is essentially a parametric method, using data from experimentally well characterised and known molecules. It is therefore used as an interpolative rather than an extrapolative technique, which cannot stray too far from "known chemistry". Thus it is not easily possible to model "kinetic control" of a reaction using the standard approach, since that requires knowledge of the transition state structure and energy. For the same reason, new molecules with unusual bonding are rarely amenable to modelling, and recourse has to be sought in the full quantum mechanical treatment of the system. Similarly, for molecular properties such as stereoelectronic effects, aromaticity, hyperconjugation and frontier orbital interactions which require a knowledge of the electron distribution within the molecule, recourse has to be made to quantum mechanical methods such as molecular orbital theory. Finally, molecular mechanics parameters are available only for certain types of bonds, and frequently are not available for many functional groups. Metal ions are also a category less easily handled at present by this type of model.

You will be using the Allinger MM2 molecular mechanics models<ref>Conformational analysis. 130. MM2. A hydrocarbon
force field utilizing V1 and V2 torsional terms {{DOI|10.1021/ja00467a001}}</ref> as implemented in the ChemBio3D program (which also supports MMFF94, useful for modelling biological systems, such as DNA, proteins, carbohydrates). MM3, MM4, Amber 11, force fields have also been developed, but are not implemented in ChemBio3D. You could alternatively use programs other than ChemBio3D, such as '''Ghemical''' (which you may remember from 2nd year), or the newer '''Avogadro'''.

'''Information Produced by the Programs''': ChemBio3D using MM2 produces an energy (in kcal mol-1) together with optimised values for bond lengths, angles etc. This energy is a rather odd quantity. It is NOT related to any thermodynamic quantity such as ΔH, and energies obtained using two different force fields CANNOT be compared. You CAN however compare two energies calculated using the same force field for two different ISOMERS. You can also calculate energy differences for simple reactions such as the hydrogenation of alkenes, particularly if this is compared across a series of related reactions. The energy itself can be dissected into contributions from the stretching ('''str'''), bending ('''bnd'''), torsion ('''tor'''), van der Waals ('''vdw''') and hydrogen bonding ('''H-Bond''') energy terms. Each term indicates the deviation from "normality" of the particular function. For example, a very positive stretch term would indicate the predicted bonds are far from the "natural" lengths, due to some geometrical feature of the molecule. Comparing these terms across say two isomers provides a natural explanation for why one isomer may be more stable than the other. Documentation for the programs being used is found here:
# Molecular Mechanics, Semi-empirical MO, Ab initio/DFT MO: [[mod:chem3d|ChemDraw/ChemBio3D]].
# Ab initio/DFT MO: [[mod:gaussview|Gaussian/Gaussview]]

===The Hydrogenation of Cyclopentadiene Dimer===

[[Image:t51.gif|right]]Cyclopentadiene dimerises to produce specifically the endo dimer '''2''' rather than the exo dimer '''1'''. Hydrogenation of this dimer proceeds to give initially one of the dihydro derivatives '''3''' or '''4'''. Only after prolonged hydrogenation is the tetrahydro derivative formed. The modelling technique here involves calculation of the geometries and energies of all four species '''1-4'''. 

The relative stabilities of the pairs of compounds '''1'''/'''2''' and '''3'''/'''4''' should indicate which of each pair is the less strained and/or hindered in a thermodynamic sense. The observed reactivity towards cyclodimerisation and hydrogenation can of course be due to either thermodynamic (''ie'' product stability) or kinetic (''ie'' transition state stability) factors. In pericyclic reactions in particular, regio and/or stereoselectivity is controlled by the electronic properties of the molecules (stereoelectronic control), and hence can only be understood in terms of ''eg'' the molecular wavefunction (''cf'' 2nd year lectures on pericyclic reactions). On the basis of the results obtained from the molecular mechanics technique you should be able to suggest whether the cyclodimerisation of cyclopentadiene and the hydrogenation of the dimer is kinetically or thermodynamically controlled.

You might wish to revisit this particular experiment in [[Mod:physical|Module 3]] of this laboratory course.

====Procedure====

Using Chem3D, define the two products '''1''' and '''2''' and optimise their geometries using the MM2 force field option. In the light of the above discussion, relate your results to the observed mode of dimerisation. The two products of hydrogenation '''3''' and '''4''' can be similarly compared so that a thermodynamic prediction of the relative ease of hydrogenation of each of the double bonds in '''2''' can be obtained. Analyse the relative contributions from the stretching (str), bending (bnd),torsion (tor), van der Waals (vdw) and hydrogen bonding (H-Bond) energy terms in terms of the relative stability of '''3''' and '''4'''.

Estimated time for completion: < 30 min.

===Stereochemistry and Reactivity of an Intermediate in the Synthesis of Taxol.===

[[Image:Taxol_molecules.gif|right]]A key intermediate '''9''' or '''10''' in the total synthesis of Taxol (an important drug in the treatment of ovarian cancers) proposed by Paquette is initially synthesised with the carbonyl group pointing either up or down. On standing, the compound apparently isomerises to the alternative carbonyl isomer. This is an example of [[organic:atropisomerism|atropisomerism]]. Clearly the stereochemistry of carbonyl addition depends on which isomer is the most stable. It is also noted that during subsequent functionalisation of the alkene, this reacted abnormally slowly!

====Procedure====

Using molecular mechanics MM2 force-field to determine the most stable isomer '''9''' or '''10''', and to rationalise why the alkene reacts slowly (hint: find literature on hyperstable alkenes!). Pay particular attention to the conformation of the resulting optimised structure, to see if any aspect of this structure could be improved by further minimisations (preceeded if necessary by a manual edit of the structure to move atoms into more correct orientations). Does the MMFF94 field produce similar results?

Estimated time for completion: < 2 hour in total.

==== Key literature ====

# S. W. Elmore and L. Paquette, ''Tetrahedron Letters'', '''1991''', 319; {{DOI|10.1016/S0040-4039(00)92617-0 10.1016/S0040-4039(00)92617-0}}
# See J. G. Vinter and H. M. R. Hoffman, ''J. Am. Chem. Soc.'', '''1974''', ''96'', 5466 ({{DOI|10.1021/ja00824a025}} {{DOI|10.1021/ja00824a025}}) and '''95''', 3051 for another nice example of atropisomerism.
# Another well known example is within Vancomycin: ''J. Am. Chem. Soc.'', '''1999''', ''121'', 3226. DOI: [http://dx.doi.org/10.1021/ja990189i 10.1021/ja990189i]
# An interesting variation is of "atropenantioselective cycloetherification" (G. ÊIslas-Gonzalez, M. ÊBois-Choussy and J. ÊZhu, ''Org. Biomol. Chem.'', '''2003''', 30-32. DOI: [http://dx.doi.org/10.1039/b208905 10.1039/b208905].
# First paper formally recognizing the new class of "hyperstable" olefins (Wilhelm F. Maier, Paul Von Rague Schleyer, ''J. Am. Chem. Soc.'', '''1981''', ''103'', 1891. DOI: [http://dx.doi.org/10.1021/ja00398a003 10.1021/ja00398a003]

==Modelling Using Semi-empirical Molecular Orbital Theory.==

In part 1, the strengths and weaknesses of a purely mechanical molecular model were illustrated. In particular, the ''endo'' stereoselectivity in Diels Alder cycloadditions was attributed to "secondary orbital" interactions, which the Molecular Mechanics approach cannot handle. In this section, such electronic aspects of reactivity will be illustrated, showing how explicit consideration of the electrons in molecules must be taken into account, and how the electrons influence bonds and derived spectroscopic properties.

===Regioselective Addition of Dichlorocarbene===

#[[Image:t55.gif|right]]'''Part 1:''' Orbital control of reactivity is illustrated in the reaction of compound '''12''' with electrophilic reagents such as dichlorocarbene or peracid. In modelling such a reaction, we require a program where the geometry of '''12''' can be predicted, and the energy of the orbitals calculated and their form displayed graphically. This experiment serves to illustrate the transition from a purely classical mechanical treatment of a molecule to a quantum mechanical treatment which includes the wave-description of the electrons. Use the ChemBio3D program and select the following methods to calculate the energy and geometry of this molecule:
## MM2 (this runs in just a few seconds and cleans the geometry up prior to applying an electronic method).
## MOPAC/PM6 or MOPAC/RM1 MO methods provide an approximate representation of the valence-electron molecular wavefunction, and in particular of the HOMO (Highest Occupied Molecular Orbital), presumed to be the most reactive towards electrophilic attack (this runs in 30 seconds or less). Does this method discriminate between the two alkene bonds? Reaction with dichlorocarbene is similar to electrophilic addition, and the HOMO indicates which of the two alkenes is the most nucleophilic.
#'''Part 2:''' The purpose here is to calculate the influence of the Cl-C bond on the vibrational frequencies of this molecule. You will compare two molecules; compound '''12''' which contains a double bond anti to the Cl-C bond and a hydrogenated version where this anti (or exo) double bond (but not the other) is replaced by a C-C single bond. The most reliable procedure for obtaining vibrations is the '''density functional''' approach (but you could use the far faster MOPAC/RM1 method and check this assertion!).
#*Using the geometry of '''12''' optimized by the previous methods, subject it and its dihydro derivative to B3LYP/6-31G(d,p) Gaussian geometry optimization and frequency calculation (see [[Mod:chem3d| instructions here]] using the keywords OPT and FREQ). Each calculation will take 1-2 hours, and is far better done using the SCAN than the laptop.
#*Load up the output LOG or FCHK (you get this from the SCAN) file (into ChemBio3D or Gaussview) and inspect any Cl-C stretching frequencies (see [[Mod:chem3d| instructions]]). Look in particular for any with a large IR intensity, and identify the two C=C stretches for the diene and the single C=C stretch for the monohydrogenated derivative. Comment on their values and any differences between the diene and the monoene. If you do spot changes, comment on whether they make sense in terms of your analysis in part one above.
#'''Part 3 (optional):''' If your curiosity has been aroused by the previous step, try modifying the substituents on the ''anti/exo'' alkene (i.e. change the =C-H group to =C-OH, =C-CN, =C-BH2, =C-SiH3) etc. Does this have any (electronic) effect on the Cl-C and C=C frequencies and is it in the expected direction? Is any effect also reproduced using MM2?
====Procedures====

#Using the Chem3D program, draw the molecule, and to perform e.g. first the MM2 step and then the MOPAC/PM6 step as above, select '''Calculations/MOPAC Interface/Minimise energy''' option from the top menus, and from the '''Job & Theory''' pane, select '''Method=PM6'''. From the Properties page, tick '''Molecular Surfaces'''. From the '''General''' pane, change the default location of the '''Results in''' folder to e.g. your documents folder (if you do not do this, the program will report that e.g. \\icfs7.cc.ic.ac.uk\yourloginname\Mopac Interface\ cannot be written to). Click '''Run''' and watch the messages at the bottom. This should take ~10 seconds (if the geometry has been pre-optimized using MM2). When this is complete, select , '''Surfaces/Molecular orbitals'''. The HOMO appears by default.
#Select '''Surfaces/Select molecular orbital''' to view the HOMO-1, the LUMO, LUMO+1 and LUMO+2. You might also want to adjust '''Surfaces/Isocontour''' to produce a more pleasing appearance for each orbital. Save each orbital as a .jpg file for insertion into your Wiki report.
#'''HINT''': Inspect the shape of your orbitals very carefully, to see if they make sense, before committing them to your report. In particular focus on whether the molecular orbital reflects the molecular symmetry of the molecule. If not, do any solutions suggest themselves? Comment on the outcome in your report.

Estimated time for completion: < 2 hour.
#Using the PM6 or RM1 optimized geometry, select the Gaussian interface and save an input file to run B3LYP/6-31G(d,p). Do the same for the monoalkene with the remaining double bond on the same side as the Cl. Save both molecules as Gaussian inputs (.gjf) files, and edit both files (using e.g. Wordpad) so that the very top line of the file shows as follows;

<pre># b3lyp/6-31G(d,p) opt freq</pre>

Any lines above this one should be deleted. Submit both jobs to the SCAN under the queue '''Chem Lab 1'''. The calculation should take <1 hour (if your starting geometry was the PM6 optimised one) to complete (provided the backlog of jobs is not high!). Do not try this on the laptop, since it will take many more hours, and the laptop may overheat! (Hint: the di-alkene can be defined as having Cs symmetry in Gaussview, and if you do this before submitting the job, the calculation will take half the time! What about the monohydrogenated system? Does that have Cs symmetry? ). When complete, download the '''Formatted checkpoint file''' for each molecule from the Web page, which should appear on your desktop. Double click this file if it does not open automatically in Gaussview 5.09, select 'open with' and navigate to Disk C/g09w/gview, and in '''Results/vibrations''' track down the Cl-C and C=C stretching vibrations in each system.

Estimated work time for completion: < 1 hour preparation time; ~4 hours elapsed time.

==== Key literature ====
#B. Halton, R. Boese and H. S. Rzepa., ''J. Chem. Soc., Perkin Trans 2'', 1992, 447. {{DOI|10.1039/P29920000447}}

=== Monosaccharide chemistry: glycosidation ===

{| cellpadding="0" cellspacing="0"

|<jmol><jmolApplet><title>Glucose02-nge</title><color>white</color>
<size>200</size><script>measure 5 8 22;zoom 150; cpk -20;</script>
<uploadedFileContents>GLUCSE02-nge.mol</uploadedFileContents>
</jmolApplet></jmol> || [[Image:anomer1.jpg|right|250px|glycosidation]]Glycosidation involves replacing the group X by reaction with a nucleophile Nu. The two sugars shown to the right give different anomers (with almost complete diastereospecificity) depending on the orientation of the OAc group on the adjacent carbon.
This effect is due to neighbouring-group-participation from the adjacent acetyl group. For the β-anomer, the intermediate oxonium cation must be attacked from the bottom face, to then allow the incoming nucleophile to replace it from the top face. Likewise, the α-anomer is formed by the acetyl oxygen attacking from the top face of the oxonium cation, allowing the nucleophile to come in from the bottom face.
|-
| <jmol><jmolApplet><title>Glucose02-nge-epi</title><color>white</color>
<size>200</size><script>measure 5 8 22;zoom 150; cpk -20;</script>
<uploadedFileContents>GLUCSE02-nge-epi.mol</uploadedFileContents>
</jmolApplet></jmol> ||[[Image:anomer2.jpg|right|430px|Neighbouring group participation]]
|}
Your task is to model these facial preferences, i.e. the diastereospecificity, using both MM2 and MOPAC/PM6 methods.
====Procedure====
#Sketch the oxonium cation rings A and B. Which group do you think is an appropriate R group to represent the chemistry while keeping the computational demand minimal? Which methods between MM2 and MOPAC/PM6 do you think is better suited for the task? Why?
#For each structure, use both methods to try to find two conformers differing in whether the acyl group is pointing '''above''' or '''below''' the plane of the oxonium cation (A and A', B and B'). Which ones of the pairs have the lower energy? Hint: use the MM2 geometry as the starting point for obtaining a PM6 geometry.
#You could try the same again but now using the intermediates C and D. You should carry out the calculation using both methods and determine the stabilisation energy by neighbouring-group-participation. Compare the results of the two methods (energy, bond lengths, bond angles, etc.). Is it still possible to get two isomers (C and C', D and D') for each?
#Can you rationalise the diastereospecificity in glycosidation using the computational results?

Estimated time for completion < 3 hours.

====Key Literature====
#D. M. Whitfield, T. Nukada, ''Carbohydr. Res.'', 2007, 342, 1291. {{DOI|10.1016/j.carres.2007.03.030}}

=: Structure based Mini project using DFT-based Molecular orbital methods =
Many (most?) reactions carried out by synthetic chemists can (and do) give mixtures of products. Often, these products are isomers (for example stereoisomers, or regioisomers which can arise from reaction at more than one site in a molecule, or from different orientations of the reacting groups). Having isolated (and ideally having separated) these isomeric products, it is important to know which isomer(s) have been formed. Often, an understanding of the mechanism of the reaction that was carried out will allow us to predict fairly confidently which isomer will predominate. You have met many key mechanistic ideas in Years 1 and 2 that enable you to do this (particularly by considering steric and electronic effects). However, it is still necessary to conclusively confirm that the expected product has been obtained, and if more than one isomer is produced, to be able to say for sure which-is-which. Sometimes, the reaction products will be solids which can be crystallised and their structures determined by X-ray crystallography, which gives detailed structural information. However, many organic compounds are oils or liquids and structure determination relies on spectroscopic methods. Mass spectrometry is a useful starting point, and high resolution mass spectrometry (HRMS) allows the molecular formula to be determined. UV and IR spectroscopy provide information on the functional groups present in the molecule. These techniques do not often allow us to distinguish between isomeric organic compounds, though. For this purpose, NMR is a primary tool because it provides information on molecular environment (chemical shifts) and connectivity (analysis of couplings between nuclei). In analysing chemical shifts, the chemist traditionally compared the observed chemical shifts to those of similar known compounds in the literature, looking for similarities to provide evidence for a structural assignment. Modern computational chemistry can provide an alternative: the 13C (also 15N,19F, 31P) spin-spin decoupled spectrum of a molecule can be predicted, often with acceptable accuracy. Computational chemistry also allows prediction of other useful spectroscopic properties, such as IR spectra and (for chiral compounds) optical rotations. For determining stereochemistry, extremely useful information can be obtained from 3-bond 1H J values, since these depend on the dihedral angle between the two protons according to the [http://en.wikipedia.org/wiki/Karplus_equation Karplus equation]. Again, molecular modelling can be used to predict dihedral angles and [http://www.stenutz.eu/conf/jhh.html hence the expected J-values].

In this experiment, you will choose a reaction from the primary literature which either is reported to give two or more isomeric reaction products, or which (based on your mechanistic understanding of the chemistry employed) has the potential to do so, even if only one isomer was reported. Ultimately, you will investigate whether 13C prediction using the GIAO approach helps in assigning the structures of the isomeric products. You might even find a paper where the original authors were uncertain as to the product structures, and be able to solve their problem! If any of the other computational techniques are useful for your example, please use them as well. The general approach to the task is as follows:
# Choose a reaction to study. You will gain extra credit (+ 5%) for finding a suitable reaction from the literature yourself. In choosing reactions, bear in mind the comment below (in the section describing how to do the 13C prediction) about the need to choose molecules that are not highly conformationally flexible. A good starting point is to look at recent issues of journals containing synthetic chemistry (e.g.[http://pubs.acs.org/journals/joceah/index.html ''Journal of Organic Chemistry''], [http://pubs.acs.org/journals/orlef7/index.html ''Organic Letters''], [http://www.rsc.org/Publishing/Journals/OB/index.asp ''Organic and Biomolecular Chemistry''], ''Tetrahedron''). Ideally, you need to choose papers which include experimental 13C NMR data for the isomeric products: this may be in the Experimental section for full papers, or in the Supporting Information which is usually available online. If you can’t (or don’t wish to) find your own reaction, some suggested examples are given below, which you may use for restricted credit. Associated with each of these are some questions to think about – even if you’ve chosen your own example, reading these may give you some idea of the kinds of question you might ask yourself when analysing your problem.
# Think about the following points and discuss them in your answer:
##How would you differentiate spectroscopically between the isomeric products? What methods would you use, and what would be the '''key''' spectroscopic differences you would look for in the spectra of the isomers in order to tell which sample is which? This is rarely discussed explicitly in synthetic papers, but the researchers doing the work will have done this as part of the research process, and thinking about how this is done is a very useful exercise for those intending on a research career!)
##Calculate the predicted 13C spectra for the isomers using the GIAO method. Include a listing of the data and assignments in your report.
##Compare your predicted data to the experimental 13C data in the paper. Do they match? Do they support the structural assignment in the paper? If not – why might this be? '''Note''': it is quite possible that your calculated data will '''not''' match those in the literature. Don’t worry! Apart from the conformational flexibility problem, there are other limitations in the computational approach. In this regard, your work is truly a research exercise – the use of this technique is at the cutting edge, and your calculations will help to determine when the method works and when it does not!
##This aspect of the course mirrors an approach commenly taken by computational and synthetic chemists. A review of the area is found here ({{DOI|10.1021/cr200106v}}) and shows in detail the way in which such problems are tackled.
# If you can, discuss the mechanism of the reaction and why the reaction shows (or doesn’t show) selectivity for one particular product isomer. Can any of the other computational techniques you’ve met in the course be used help to explain the selectivity?

== Objectives and Suggested Structural Explorations ==

You will be given full credit for attempting '''ONE''' of the below (and 5% bonus for one of your own devising not listed below). This part of the module should take < 10 hours spread out over two days to complete.
==== Stereoselective dissolving metal reductions ====
[[image:reduct1.jpg|right]]You met the use of dissolving metals (e.g. Li/NH3) for reduction of carbonyl compounds and aromatic rings (Birch reduction) in Year 2 (Functional Group Interconversions course). In a recent natural product total synthesis ({{DOI|10.1016/j.tet.2006.12.019}}), these conditions were used for stereoselective reduction of a cyclic ketone to an alcohol: ketone '''5''' was reduced with complete stereoselectivity to give alcohol '''6''' (Scheme 1 in the paper). How would you tell the reaction had worked, and which stereoisomer had formed? Do the predicted 13C (and 3JH-H) data match the reported? Why was this stereoisomer obtained? The optical rotation of '''6''' is reported. Does it match calculation? (you may have to play with the orientation of both the OH and propenyl groups to get a good match).

=== Regio- and stereoselective conversion of alkenes to epoxides ===
[[image:epoxide.jpg|right|thumb]]As you know from Years 1 and 2 (especially Year 2 Functional Group Interconversions course), epoxides are highly versatile synthetic intermediates because they undergo ring opening with a wide range of nucleophiles. In a recent paper ({{DOI|10.1016/j.tetasy.2005.02.012}}), it was shown (Scheme 6 in the paper) that a 1,3-diene (compound '''13''') can be regioselectively epoxidized to give either stereoisomer '''14''' or '''15''' depending on the reaction conditions used. How would you tell that the “correct” alkene had been epoxidised? Unfortunately the authors don’t include 13C data for their products in the paper, but they are available in the literature (do a search on Beilstein). Do the products’ 13C data match your calculations? Why do the two sets of epoxidation conditions give different stereoisomers?

=== Assigning regioisomers in "Click Chemistry" ===
[[Image:Click.gif|left]]The 1,3-dipolar cycloaddition between an azide and an alkyne to give a 1,2,3-triazole has been known for a long time. However, in 2002, two groups reported independently that the use of a Cu(I)-catalyst greatly speeds up the reaction. So facile is the catalysed chemistry that it is now often called the “click reaction”, being a classic example of the philosophy of “[http://en.wikipedia.org/wiki/Click_chemistry click chemistry]” introduced by Nobel Laureate K. Barry Sharpless, which aims to exploit reactions which “are tailored to generate substances quickly and reliably by joining small units together”. The ease and high selectivity of the click reaction has led to its widespread use in important fields such as materials science and biology – for example, the specific introduction of tags and labels into proteins.
When substituted alkynes and azides are used, there are two possible regioisomeric products, '''A''' and '''B'''. Under Cu(I) catalysis, the 1,4-isomer '''A''' predominates; interestingly, however, Ru(II)-catalysed reactions give mostly the 1,5-isomer '''B''' instead. How can we tell these apart spectroscopically? Look at the original paper (''J. Am. Chem. Soc.'' '''2005''', ''127'', 15998; {{DOI|10.1021/ja054114s}}). For the example in Table 1, entry 1, calculate the 13C spectra for the depicted product isomer and for the alternative regioisomer. Which best matches the data reported in the paper? For comparison, can you find data for the other isomer anywhere in the literature?

=== Investigating the regioselectivity of the Baeyer-Villiger reaction ===
[[image:bv2a.jpg|left|thumb]][[image:bva1.jpg|right]]The [http://en.wikipedia.org/wiki/Baeyer-Villiger_oxidation Baeyer-Villiger reaction] converts ketones into esters, effectively inserting an oxygen atom between the carbonyl group and the alpha-carbon. It is usually carried out using mCPBA. For unsymmetrical ketones, there are two possible regioisomeric products; usually it is possible to predict which isomer will predominate in line with the migratory aptitude of the ketone substituents. Investigate '''ONE''' of two recent literature examples where the reasons for the regiochemical outcome are not so obvious:
*The Baeyer-Villiger reaction was used ({{DOI|10.1021/jo030377y}}) in a recent synthesis of analogues of beta-lactam antibiotics. In Scheme 5, reaction of '''10d''' gives '''11d''' and '''12d'''. Are the regiochemical assignments correct, and why is the regioselectivity low in this case?
*During a total synthesis of the natural product (-)-kainic acid ({{DOI|10.1016/S0040-4020(02)00379-4}}), a neuropharmacological tool, a highly regioselective Baeyer-Villiger reaction was a key step (conversion of compound '''14''' into '''15'''). Do the predicted 13C data fit with the reported ones? Why is this reaction so regioselective?

=== The total synthesis of (-)Cubebol ===

[[Image:cubebol.gif|right]] Cubebol is a natural sesquiterpene alcohol with a cooling and refreshing taste. It was patented as a cooling agent (i.e. to make you feel you are eating ice-cream without actually having to freeze the ingredients!). The total synthesis has recently been reported: {{DOI|10.1021/jo9022974}} in which much spectroscopic information about the molecule is available. Your project will involve testing that this information and the assignments are correct for the absolute configuration shown.

=== General Reference===
K. Mori, ''The Chemical Record'', '''2005''', ii5'', 1-16. {{DOI|10.1002/tcr.20030}}

== Relevant computational techniques for Mini-project ==

===Predicting the 13C NMR Spectrum of a compound===

The 13C (also 15N,1NF, 31P) spin-spin decoupled spectrum of a molecule can be predicted using two quite different methods.
# The first is a rule-based approach is derived from a fragment library, and is applicable mostly for organic molecules. The advantage is that the prediction is extremely rapid, and fairly general. The downside is that the accuracy is only around 3-5 ppm, and does not take into account local conformations, differential solvation of different groups, etc. It is not applicable for many organometallic and inorganic systems.
# The second is the so-called GIAO approach using quantum mechanical density functional theory. The background to this, and a famous recent example can be found in the article by Rychnovsky1 on a revision of the structure of '''Hexacyclinol''' ( {{DOI|10.1021/ol0611346}}). He reports that the mean error for the 23 carbon shifts in the predicted structure was around ± 1.8 ppm, with a maximum error of around 5.8 ppm. An improved procedure which reduces the mean and maximum errors by one half will be used here2 ({{DOI|10.1021/np0705918}}), although a number of caveats for successful prediction should be noted. The most serious is that the method is '''highly''' sensitive to the conformation of the molecule. If various different conformations are possible (and for some molecules, 100s of reasonable conformations can sometimes be imagined), they should all be scanned by this method. Since this is clearly not feasible in a reasonable time, you should not choose a problem that has conformational ambiguity.

====Procedure====

=====Creating an initial Molecule input file =====
You will need to sketch your molecule in ChemDrawPro+ChemBio3D/Gaussview and perform an initial refinement of its 3D geometry using MM2. If it contains only simple elements (CHNO, Si, P, S, halogens) then the chances are that a molecular mechanics refinement will be possible. At this stage, whilst the calculations still take only a few seconds, you might wish to investigate several conformational possibilities to see which might be the lowest (but don't try more than say 5). Some conformations can be preset (a worthwhile one is to always try to get 6-membered rings into a chair, and e.g. esters R-CO-O-R' oriented such that the R-C bond is antiperiplanar to the O-R' bond). If the mechanics procedure fails because of lack of parameters, try eg the MOPAC/AM1 approach instead. If both of these fail, try the Gaussian procedure, using the HF (Hartree-Fock) method and an STO-3G basis set. This initial geometry will then have to be '''refined/optimized''' using the following method.
# In ChemBio3D, go to Calculations/Gaussian/Create Input File.
# Select Job Type/Minimise; Method DFT=mpw1pw91
#*'''A note on DFT methods:'''. A wide variety of DFT methods have been proposed {{DOI|10.1021/jp710179r}}. This particular one has been extensively tested for the calculation of NMR properties and found to be somewhat superior to the more common B3LYP procedure.
# The Basis set to be set to 6-31G(d,p)
#*'''A note on basis sets:''' A wide variety of basis sets have been [https://bse.pnl.gov/bse/portal described] (including a set optimized specifically for NMR shift calculations, {{DOI|10.1021/ct800013z}}, type '''pcS''' into the search box [https://bse.pnl.gov/bse/portal here]) and often a decision on which basis set is most appropriate for which property being computed has to be made. In the case of NMR as a property, it is not necessarily true that the larger basis set is the better one! You will revisit basis sets in modules 2 and 3 of this course.
# Save the resulting file to your H: drive, making sure it is saved as a '''Gaussian Input file''', with the suffix '''.gjf'''.
# Find the file in Windows Explorer, and with a right-mouse-click, open it with the WordPad program.
# Delete all lines at the top, leaving only the following line, which should be edited to show something like the following
<pre># mpw1pw91/6-31g(d,p) opt(maxcycle=25)

Geometry optimization for literature compound

0 1
atom1-symbol 0 x-cooordinate of atom1 y-cooordinate of atom1 z-cooordinate of atom1
atom2-symbol 0 x-cooordinate of atom2 y-cooordinate of atom2 z-cooordinate of atom2
... ... ...</pre>This shows the keyword line at the top, a blank line, a title card, another blank line, a charge/spin card (we will assume that your unknown is neutral, i.e. charge=0 and a singlet spin state, i.e. spin=1) and the first line of atom coordinates. If you need to calculate a charged species, change the '''0''' to e.g. '''-1''' (for an anionic species). Whilst you are at it, check to see if your coordinates have any atom type designated '''Lp'''. If any such lines are present, delete the entire line. Lp is a Lone-pair, and is sometimes added by the Molecular Mechanics part of the program. However, if Gaussian sees it, it gets very confused, and will not run at all!. The keyword value maxcycle=25 is because sometimes the geometry optimization meanders very close to convergence, and this limits this meander. Re-save this file, making sure you save it as '''TEXT''' and ''NOT'' RTF and that it retains the suffix '''.gjf'''.

==== Submitting this file to the SCAN for geometry optimization ====
#[[Image:scan2.jpg|thumb|left]] [[Image:scan1.jpg|thumb|right]] Go to [https://scanweb.cc.imperial.ac.uk/uportal2/ the SCAN Webpage], log in, and first create a project (it could simply be called Mini-project). Then, '''New Job/Chemistry Lab 1''', then select Gaussian/Your project, and finally the name of the Gaussian input file you have just saved, along with a descriptive title.
#You can view your job list, when a display of the type shown below should appear: Jobs in the '''Chemistry Lab 1''' pool also run during the day, but with a r concurrency of 8. When there are many jobs you may have to wait overnight for yours to finish. The status of the pools can be inspected by selecting Pools from the menu on the left: If your suspected molecule is large (more than about 30 non-hydrogen atoms) it may require more than about 9 hours of '''wall''' time. If the job returns no output overnight, it may well have run out of time (it has about 9 hours in which to complete the calculation). The '''Chemistry Lab 1''' will run the job to completion, but you may have to wait a while for it to start running in the first place.
# [[Image:scan3.jpg|thumb|right]]When the job shows as Finished, select the Gaussian Checkpoint file as the required output and download it (probably to the desktop, or wherever the browser tells you). Double-click the file to open Gaussview (it may happen automatically) and check that the optimised geometry is still reasonable. Invoke '''File/Save as''' and replace the original Gaussian input file you created with Chem3D. It now has a fully optimised geometry at the mpw1pw91/6-31(d,p) level, rather than the initial sketch of before.

==== Troubleshooting ====
# If the system responds that the formatted checkpoint file '''does not exist''' its quite probable that the calculation failed. Try instead to download the Log file, which may have error messages that help you diagnose what has gone wrong. Two common reasons for the failure are
## There was an error in the input .gjf file. A common error is the positioning or omission of blank lines. Check with the above to ensure they are correctly positioned. Another error is that the keywords are mis-typed. Gaussian will fail for either reason, but it should put out an error message in the log file.
##The best way to eradicate syntax errors before submission to SCAN is to run Gaussian on your laptop for a few seconds at least, this being better than waiting up to 24 hours to find that a trivial error stopped the calculation. You can run a Gaussian input either from ChemBio3D, or Gaussview.
## The calculation may have run for 9 hours and then run out of time. This means that the molecule may be rather large (> 30 non hydrogen atoms), or very conformationally mobile. You could try resubmitting with maxcycles set to something lower.

==== Submitting this file to the SCAN for NMR Chemical Shift calculation ====
# Having created a new .gjf file containing the optimized geometry, repeat the Wordpad editing procedure as described above, but this time ensure the top line of your .gjf file contains the following (if the literature reports a different solvent, replace chloroform with that solvent): (a list of defined solvents is available at the bottom of [http://www.gaussian.com/g_tech/g_ur/k_scrf.htm this page])
<pre># mpw1pw91/6-31(d,p) NMR scrf(cpcm,solvent=chloroform)

NMR calculation for literature compound

0 1
atom1_symbol 0 optimized-x-cooordinate of atom1 optimized-cooordinate of atom2 optimized-cooordinate of atom3</pre>
Resubmit this new input file for calculation as described above. This will take much less time to calculate than before. Whilst it is possible to catenate the two jobs you have run (the optimization and the NMR calculation), this requires a great deal more editing of the .gjf file, and hence scope for errors.

==== Analyzing the NMR Chemical Shift calculation ====
#[[Image:scan4.jpg|right|thumb]] When this second calculation is finished, download this time the Gaussian Log file (instead of the checkpoint file). Open this in Gaussview and from that program, select '''Results/NMR''' (if the NMR keyword is greyed out, it means the calculations was not in fact successful).
# From the Spectral display that appears, select the '''C''' nucleus, and the appropriate Reference Value. Click on any peak to find out what its chemical shift is, and compare with the spectrumreported in the literature.
# You should note that carbons attached to "heavy" elements (particularly eg halogens) have shifts which need correction for so-called Spin-orbit coupling errors. Typically, C-Cl needs correcting by -3 ppm, C-Br by -12 ppm, and C-I by about -28 ppm. First row transition metals are around -3ppm. Other elements to be determined!2. Another systematic error present is that the carbonyl of esters, amides etc tends to be out by about 5ppm. Use the following simple correction for such carbons only: δcorr = 0.96δcalc + 12.2.
# You can probably use your calculation to actually assign the 13C shifts to the carbons of your molecule. If you spot one or more carbons out by more than about 5ppm, its quite likely that you have the wrong conformation of your molecule in that region (i.e. the method can actually be used for conformational analysis), or of course that the original assigment in the literature is wrong. This actually happens quite often!
# The method should work for other nuclei (except hydrogen, which requires much greater accuracies to be really useful). First row transition metals (organometallics) appear to be reasonably handled.
# Complete this section by returning to [https://scanweb.cc.imperial.ac.uk/uportal2/ the SCAN portal] and click on the '''publish''' link next to the job that carries the NMR prediction. This will deposit your calculation into a so-called '''Digital repository'''. Quote the entry in your Wiki pages as <nowiki>{{DOI|10042/to-xyz}}</nowiki> where xyz is the entry generated by the previous operation.

====References====

# S. D. Rychnovsky, ''Org. Lett.,'' '''2006''', ''13'', 2895-2898. {{DOI|10.1021/ol0611346}}
# C. Braddock and H. S. Rzepa, ''J. Nat. Prod.,'' '''2008''', ''71'', 728-730. {{DOI|10.1021/np0705918}}
# A recent development is an enhanced technique for accurately computing 1H chemical shifts: {{DOI|10.1021/jo900482q}}
# Goodman has produced some interesting tools for aiding NMR analysis.
## {{DOI|10.1021/jo900408d}}
##[http://www-jmg.ch.cam.ac.uk/tools/nmr/ Applet]
## {{DOI|10.1021/ja105035r}}
## [http://www.spectroscopynow.com/coi/cda/detail.cda?id=24215&type=Feature&chId=5&page=1 Blog commentary]

===Predicting the 3J H-H couplings of your compound===

The above technique is reliable for 13C shifts, but less so for 1H shifts. However, three-bond couplings can be predicted reasonably well using a very rapid and simple method based on the Karplus equations. To do this, you will need to have a 3D model of your unknown, which should emerge out of your 13C prediction in the preceeding section. The start point should be to use Chem3D to save an MDL Molfile of your final coordinates. This can then be read into [http://www.ch.ic.ac.uk/local/organic/janocchio/index.html Janocchio,] to provide the coupling constants.

It is also possible to compute J couplings using quantum mechanical methods (Gaussian keyword '''NMR(spinspin)''') but these calculations are '''highly''' time consuming if accurate results are to be obtained.

====Reference ====

D. A. Evans, M. J. Bodkin, S. R. Baker, G. J. Sharman, ''J. Magn. Reson,'', '''2007'''. * {{DOI|10.1002/mrc.2016}}

=== Predicting the IR Spectrum of a compound ===

Calculating the vibrational normal modes of a molecule is simpler than NMR since it can be done in a single job, but it may also be more time consuming. Only try this for smaller molecules (<25 non hydrogen atoms, including the first row transition metal series).
==== Procedure ====
Create an initial .gjf file, and modify it thus;
<pre># b3lyp/6-31G(d,p) opt freq

Geometry optimization and vibrational frequencies for literature compound

0 1
atom1-symbol 0 x-cooordinate of atom1 2-cooordinate of atom2 -cooordinate of atom3
atom2-symbol 0 x-cooordinate of atom1 2-cooordinate of atom2 -cooordinate of atom3
... ... ...</pre>
In the resulting output, <pre>Sum of electronic and thermal Free Energies= -3170.440313</pre>
gives you in effect ΔG = ΔH - T.ΔS. You can use this term to compare the '''difference''' in free energies between two molecules, remembering that it is expressed in Hartrees; 1 Hartree = 627.5 kcal/mol.
==== Analyzing the Vibrational Spectrum ====

Download the.fchk file from the SCAN page, and by double-clicking, open it in Gaussview. From '''Results/Vibrations''', select the '''Save normal modes''' from the '''Run FreqChk''' pop-up box and inspect the normal modes and their predicted intensities, using the animation feature to help describe them. Errors in the predicted wavenumbers are systematically too high for stretches (which means they can be corrected using empirical factors) by around 8%; bending and lower frequency modes are normally about right. Pay particular attention to the predicted intensities, which may help you to assign the vibrations. If you get any apparently negative modes, you will have in fact obtained a transition state (or higher order) stationary point.

==== Reference ====

R. Janoschek, ''Pure and Applied Chemistry'', '''2001''', ''73'', 1521-1553. {{DOI|10.1351/pac200173091521}}

=== Predicting the Optical Rotation (OR) and the Electronic Circular Dichroism (CD/UV-Visible) Spectrum of a compound ===
Measuring optical rotations is one of the oldest spectroscopic techniques, dating back well into the 19th century, and a mainstay of organic chemistry until IR/NMR etc came along mid 20th century. Although the theory of how molecules interact with polarized light has been known for a long time, it is only in the last 5 years or so that computers have become sufficiently fast to solve the problem to the required accuracy, which in fact comes in two parts. The simpler is to see if the absolute sign of the optical rotation predicted for a given absolute configuration of a molecule corresponds to that measured. Because the sign can easily change as a result of apparently minor changes to the structure of the molecule (or even in extreme cases, its conformation), there is little ''intuition'' that can be applied, or indeed simple rules. A full quantum mechanical calculation is pretty much the only reliable method for predicting the absolute sign of the OR. The second aspect is predicting the magnitude of the rotation. This again can vary from close to zero, to many thousands! It is generally accepted that only compounds with ORs of magnitude >|100| (or at a pinch >50) can be successfully used to predict absolute configurations with near total confidence. So you should only attempt to predict the OR of an asymmetric molecule if it fulfills these criteria. Another chiro-optical property is the CD spectrum. This is essentially the UV spectrum of the molecule, with the difference that it is recorded with '''chiral''' light. The two enantiomers of a disymmetric molecule interact differently with this light (think of it as opto-electronic diastereomers), and particularly the sign of the intensity of each electron transition can be either positive or negative. The resulting CD spectra are exact mirror images of each other for each enantiomer of the molecule, which means that distinguishing between them is trivial. The crucial difference between CD and OR is that the former is very much less sensitive to conformation, and hence the answer so much more definitive. There are other types of '''chiro-optical''' spectroscopies (Vibrational circular dichroism, Raman Optical Activity) which can be even more definitive, but these are still rarely used.

Proceed as follows:
==== Procedure for Optical Rotation====
Take the output of the previous frequency or NMR calculation (i.e. the optimized geometry), and run a job of the following type:

<pre>
# cam-b3lyp/ 6-311G(d,p) polar(optrot) scrf(IEFPCM,solvent=chloroform) CPHF=RdFreq
This is a blank line; put no text in it
Optical rotation for literature compound
This is a blank line; put no text in it
0 1
firstatom-symbol 0 x-cooordinate of atom1 2-cooordinate of atom1 -cooordinate of atom1
...
lastatom-symbol 0 x-cooordinate of lastatom 2-cooordinate of lastatom -cooordinate of lastatom
This is a blank line; put no text in it
589nm
This is a blank line; put no text in it
</pre>
The Cambridge variation on the B3LYP density functional method is used, which improves the prediction of chiro-optical properties compared to the normal B3LYP version. The keyword '''aug-cc-pvdz''' invokes an augmented (i.e. with additional diffuse basis functions), correlation-consistent double-ζ valence polarization basis set suitable for optical rotation calculations and polar(optrot) calculates the [alpha]D optical rotation components of an asymmetric mol. The i wavelength of the incident light (589nm is the sodium D line) is read in using the keyword CPHF=RdFreq and the line appended as 589nm after a blank line following the coordinates. A final blank line follows the frequency line. Specify the appropriate solvent in the SCRF keyword. If this recipe fails to converge (it can do occasionally), try instead a faster and simpler basis set to replace the aug-cc-pvdz one: '''6-31G(d,p)'''.

In the resulting output, e.g. [ALPHA] ( 5890.0 A) = -324.5 deg gives the estimated optical rotation for the exact enantiomer that you built (try submitting the other enantiomer and see if you get the opposite rotation). The value given by [Alpha]D is misleading, it being the non-frequency dependent approximation to this parameter (and regarded as less accurate than the frequency dependent value at 589nm which follows). The method will reliably predict whether the optical rotation corresponds to the enantiomer you have built if [Alpha]D > 100°, but becomes increasingly unreliable for lower values. The OR is also highly sensitive to conformation; even a 60° rotation of an OH group can alter its value by a factor of two! Turned on its head, predicting OR could be regarded as a highly sensitive method for conformational analysis! You should be aware that this calculation an be quite time consuming, and molecules with > 20 non hydrogen atoms should not be attempted.

==== Procedure for the CD (Circular Dichroism) Spectrum ====

Use the following keywords, which invokes the so-called time-dependent DFT method, where the first 20 electronic singlet excitations are included (you can reduce this to a much smaller value, eg 3 or 5, or a much higher one if you want to simulate the high energy/UV region of the spectrum).

<pre># cam-b3lyp/6-311G(d,p) td(NStates=20) scrf(IEFPCM,solvent=chloroform)

Circular dichroism for literature compound

0 1
atom1-symbol 0 x-cooordinate of atom1 2-cooordinate of atom2 -cooordinate of atom3
atom2-symbol 0 x-cooordinate of atom1 2-cooordinate of atom2 -cooordinate of atom3
... ... ...</pre>

The spectrum can be viewed using the ''Results/UV-Vis'' option in Gaussview 3 or 5. The regular UV spectrum is shown first, followed by the CD version.

==== References ====

# P. J. Stephens et al, ''Chirality'', '''2008''', ''20'', 454-470. {{DOI|10.1002/chir.20466}} {{DOI|10.1063/1.1477925}} {{DOI|10.1063/1.1436466}}
# B. Mennucci, M. Claps, A. Evidente, and C. Rosini, ''J. Org. Chem.,'' '''2007''', ''72'', 6680-6691. {{DOI|10.1021/jo070806i}}
# For an example of a calculation, see that for pentahelicene ([alpha]D 2061°) {{DOI|10042/to-888}} If you really want to entertain yourself, try something larger such as decahelicene!
# For a recent application to another type of highly chiral molecule, see {{DOI|10.1021/ol901172g}}

===Using the SCAN for Density functional MO calculations===

You can use the '''SCAN''' to run a Gaussian calculation. Using ChemBio3D, pre-optimise the structure using a fast method such as Molecular mechanics before submitting the DFT calculation. If you do not do this, the latter will take much longer! Create a Gaussian input file (a '''.gjf''' file) as [[mod:chem3d|described]] in the instructions, and then
# Go to [https://scanweb.cc.imperial.ac.uk/uportal2/ https://scanweb.cc.imperial.ac.uk/uportal2/] and log in.
# Select '''Projects''' and '''create a project name''' suitable for your needs.
# Select '''New Job''', then:
##'''Chemistry lab 1''' (a queue of width 8 which runs during the day)

# Then enter your project, and upload the .gjf or .com file. Put in a descriptive title to help remind you of the purpose of the calculation, and click on submit. The Job will show as either pending (for all overnight jobs) or running (for the first 8 jobs on Chemistry Lab 1). When complete, it can be collected from the same web page. In particular, if you select the '''Formatted checkpoint file''' from output list, and download it, Gaussview will open it and display the result of your calculation. You can also open this file with ChemBio3D.
# The SCAN is powerful enough that if you wished, all the molecules in this section could be submitted using the Gaussian program. You can submit multiple jobs, one after another using this technique. You could also increase the level of theory. In this case, change the basis set from 6-31G(d) to e.g. cc-pVTZ, or you could e.g. include a vibrational analysis ('''freq''' keyword) which in fact will result in an entropy correction to the energy, to give in effect a ΔG for your energy (this energy is labelled <tt>Sum of electronic and thermal Free Energies=</tt>)

==== Troubleshooting the SCAN outputs ====

The following lists some of the things that might go wrong, and what to do about them. If you identify a reproducible cause of failure yourself, please feel free to add to the list below!
* A job is finished but it returns no formatted checkpoint file. It is likely that there was an error in the input .gjf file. A common error is the positioning or omission of blank lines in this file or that one of the keywords is mis-typed. Another error is that a keyword may be repeated (thus Gaussian does not much like repetition of the '''opt''' keyword). Download the log file (if it exists) and open it with eg '''Wordpad'''. Check that blanks lines are all correctly present and positioned and for keyword errors or duplication. The output may give a clue of sorts, but the presence or absence of blank lines often confuse it. The below is an example of how an unrecognized keyword is flagged.
<pre>
# b3lyp/6-31G(d) nopt
---------------------
QPERR --- A SYNTAX ERROR WAS DETECTED IN THE INPUT LINE.
# b3lyp/6-31G(d) nopt
^</pre>
*If you cannot get a log file from the finished job, it is likely it ran out of time (each job has a limit of 48 hours). Put simply, your molecule (or the property you are trying to calculate) is a tad too big/demanding!
*It is important if a job fails, to provide as much evidence as you can to demonstrators. Thus at a minimum, you should have to hand the input file (.gjf), and ascertain if running it produces any output. Do also remember that computers are relatively reproducible. If a job fails, resubmitting it will most likely produce a second failure. Rather than simply resubmitting a job, you '''must''' resolve the undoubted error the input contains. Remember that errors can be caused by what is called '''white space''' (which of course since it consists of nothing much, is easily disregarded), and that often even experienced demonstrators might fail to spot that extra bit of white space that is causing the error. If nothing obvious strikes you about an input, it might be easier to throw it away and start again rather than wait eg 24 hours to find it has (reproducibly) failed again!
*It is also a good idea to '''run Gaussian on your laptop for a few seconds at least, this being better than waiting up to 24 hours to find that a trivial error stopped the calculation.''' You can run a Gaussian input either from ChemBio3D, or Gaussview (by now you will appreciate that Gaussian itself is really not very good at handling and describing errors).

= Help =

#In addition to demonstrators and staff, you may wish to keep an eye out on the [[Mod:latebreak|late breaking news]] page for general updates, and the discussion areas for [[Talk:Mod:organic|Module 1]], [[Talk:Mod:inorganic|module 2]] and [[Talk:Mod:physical|module 3]] where people document their experiences, suggest bug fixes etc.

= General References =

<references/>

= Module 1 Marking =

The marks for this module are split 60% for the four components of part '''1.2''' and '''1.3''' (15% each) and 40% for the miniproject part '''2'''. You will be assessed not simply on whether you got the ''right'' answer, but on your analysis of the problem, how you might have designed control calculations, or e.g. worked out ways of making the modelling more efficient. The project is at least as much about how you go about organising your '''workflows''' in the time you have decided to spend on it, as it is about getting the right answer. Marks for each individual component will also be awarded for how you cite and quote the literature (in particular for citing any relevant references that we do '''not''' give you in the notes). Remember, being critical is more important than merely reproducing quotes from an article. After all, the original people who reported the chemistry may have not understood what happened themselves, and it is perfectly possible that you may actually be able to critically improve that understanding!

Your grade will be recorded in Blackboard and comments on your experiment will appear in the discussion section of your Wiki report. If you want to discuss your experiment and its grade, please contact Prof Alan Armstrong directly or Dr Bao Nguyen directly.

----
See also: [[Mod:timetable|Timetable]],[[Mod:lectures|Intro lecture]],[[mod:laptop|Laptop use]], [[mod:programs|Programs]], [[mod:organic|Module 1]], [[Mod:inorganic|Module 2]], [[Mod:phys3|Module 3]],[[Mod:writeup|Writing up]]

Talk:Mod 1: Celeste van den Bosch

2011-10-27T10:43:35Z

Mjhughes:

===First week feedback===
Overall it's hard to give a good assessment of your work, mainly because there's no indication as to which answer is completed. Consequently, I'm giving you the assessment of the page 'as is' on Friday night. Some of the comments probably will be addressed by you in due course, if not already.

*The presentation of the data is good and easy to follow/mark to me. I like to see a reaction/structure scheme with numbering for each questions so I don't have to flip between pages.
*You may find that a few of your pictures could be too big, depending on the width of the browser. This is due to the fact that wiki engine doesn't scale pictures along with text. An example of a code which will help you is shown below, in which the width of your picture is 500 pixels.
<pre style="color:blue"> [[Image:filename.jpg|alignment(right/left/middle)|500px|description]] </pre>
*Question 1 - DiCy: Splitting Table 1 may help. You'll need to provide an analysis (what we really want to see!) to compare 1 with 2 and 3 with 4. Please pay attention to the source of the difference in energy, and if possible identify the part of the molecules responsible for it.
*Question 2 - Taxol: You'll need to rationalise the stability of the double bond. Pictures or Jmol to illustrate the structures in Table 2 is crucial here. MMFF94 calculation is still missing as is an explanation for the unusually stable double bond. You did notice that it's at the bridgehead and is normally strained though.
*Question 3 - MOPAC: Do we need LUMO+2? Your MO pictures looks OK, but a top-down point-of-view will probably be better. Most of your analysis is yet to be included, but what's the structure of the monoene?
*References: you've two in there. Obviously as you fill in the details and analysis, you'll need references to back your statements up.

===FEEDBACK (After Week 2)===

Q1: The energy values are all correct and the discussion of selectivity in the Diels-Alder reaction is good. It is not really possible to say whether the hydrogenation reaction is under thermodynamic or kinetic control; the kinetic product (lowest energy transition state) could be the same as the thermodynamic product (lowest energy product) and in fact this is usually the case – the Diels-Alder reaction is a bit of an odd case. So if you know the product is the higher energy molecule you can say for sure that the reaction is under kinetic control since it can’t be under thermodynamic control, but if the lower energy molecule is formed you still can’t say without further analysis (transition state calculations) and more information (reaction conditions etc). As a side note: most reductions of this sort are metal-catalysed, irreversible, and therefore under kinetic control. Your comparison of the contributions to strain correctly shows the major factors (the % analysis would have been better if it had shown difference as a % of total energy – or more simply the absolute difference in kcal). The reason bending strain is worse in 3 than in 4 is because the bond angles deviate more from the sp2 ideal in that compound – due to the constraints of having a double bond in the bicyclic half of the molecule.

Q2. Your energies are all good and the favoured isomer is correctly identified. You have a brief comment on the preference for the chair conformation for the 6-ring; a more complete analysis would also show the higher energy twist-boat for compound 9; you could also have briefly described the other ways in which you attempted structural optimisation. Hyperstable alkenes are correctly defined.

Q3. The MOs look good and the IR stretch frequencies are correct. The regioselectivity in the reaction with dichlorocarbene is well explained. The difference in the C-Cl stretching frequency when the double bond is removed is due to a pi-sigma* orbital interaction. When the double bond is removed it can no longer donate into the C-Cl antibonding orbital, so the bond is stronger.

Q4. The energy values and structures all look good. R=methyl is the correct choice here for a generic alkyl group and as you have discussed semi-empirical methods are best for this case. Your finding that A/B/C/D are lower in energy than A’/B’/C’/D’ is right. As your values show, for PM6 calculations, A=C and B=D; the calculation method is capable of incorporating the neighbouring group effect into a hybrid structure which is treated as a non-classical carbocation. The stereoselectivity in the reaction is due to two favourable effects, A/B are both lower in energy and more reactive (better orientation and trajectory for nucleophilic attack) than A’/B’.

MP. This looks like a good mini-project – a mixture of diastereomers with rigid looking structures. The NMR analysis is good and looking at different conformations is a good idea; if the energies of these conformations have been obtained, it should even be possible to look at the Boltzmann distribution at various temperatures and work out the expected contribution to physical properties of each. Another aspect that could have been looked at here is the RMS you get when comparing calculated values for the wrong isomer to lit data; if the wrong isomer has a significantly worse correlation to the data than the correct isomer this supports the use of predicted NMR as an analytical technique in this case. The optical rotation calculations seem to be good at getting the appropriate sign of the isomer if not the absolute value; in some cases this could be used to distinguish between isomeric products.

Talk:Mod 1: Celeste van den Bosch

2011-10-27T10:43:16Z

Mjhughes:

===First week feedback===
Overall it's hard to give a good assessment of your work, mainly because there's no indication as to which answer is completed. Consequently, I'm giving you the assessment of the page 'as is' on Friday night. Some of the comments probably will be addressed by you in due course, if not already.

*The presentation of the data is good and easy to follow/mark to me. I like to see a reaction/structure scheme with numbering for each questions so I don't have to flip between pages.
*You may find that a few of your pictures could be too big, depending on the width of the browser. This is due to the fact that wiki engine doesn't scale pictures along with text. An example of a code which will help you is shown below, in which the width of your picture is 500 pixels.
<pre style="color:blue"> [[Image:filename.jpg|alignment(right/left/middle)|500px|description]] </pre>
*Question 1 - DiCy: Splitting Table 1 may help. You'll need to provide an analysis (what we really want to see!) to compare 1 with 2 and 3 with 4. Please pay attention to the source of the difference in energy, and if possible identify the part of the molecules responsible for it.
*Question 2 - Taxol: You'll need to rationalise the stability of the double bond. Pictures or Jmol to illustrate the structures in Table 2 is crucial here. MMFF94 calculation is still missing as is an explanation for the unusually stable double bond. You did notice that it's at the bridgehead and is normally strained though.
*Question 3 - MOPAC: Do we need LUMO+2? Your MO pictures looks OK, but a top-down point-of-view will probably be better. Most of your analysis is yet to be included, but what's the structure of the monoene?
*References: you've two in there. Obviously as you fill in the details and analysis, you'll need references to back your statements up.

FEEDBACK (After Week 2)===First week feedback===

Q1: The energy values are all correct and the discussion of selectivity in the Diels-Alder reaction is good. It is not really possible to say whether the hydrogenation reaction is under thermodynamic or kinetic control; the kinetic product (lowest energy transition state) could be the same as the thermodynamic product (lowest energy product) and in fact this is usually the case – the Diels-Alder reaction is a bit of an odd case. So if you know the product is the higher energy molecule you can say for sure that the reaction is under kinetic control since it can’t be under thermodynamic control, but if the lower energy molecule is formed you still can’t say without further analysis (transition state calculations) and more information (reaction conditions etc). As a side note: most reductions of this sort are metal-catalysed, irreversible, and therefore under kinetic control. Your comparison of the contributions to strain correctly shows the major factors (the % analysis would have been better if it had shown difference as a % of total energy – or more simply the absolute difference in kcal). The reason bending strain is worse in 3 than in 4 is because the bond angles deviate more from the sp2 ideal in that compound – due to the constraints of having a double bond in the bicyclic half of the molecule.

Q2. Your energies are all good and the favoured isomer is correctly identified. You have a brief comment on the preference for the chair conformation for the 6-ring; a more complete analysis would also show the higher energy twist-boat for compound 9; you could also have briefly described the other ways in which you attempted structural optimisation. Hyperstable alkenes are correctly defined.

Q3. The MOs look good and the IR stretch frequencies are correct. The regioselectivity in the reaction with dichlorocarbene is well explained. The difference in the C-Cl stretching frequency when the double bond is removed is due to a pi-sigma* orbital interaction. When the double bond is removed it can no longer donate into the C-Cl antibonding orbital, so the bond is stronger.

Q4. The energy values and structures all look good. R=methyl is the correct choice here for a generic alkyl group and as you have discussed semi-empirical methods are best for this case. Your finding that A/B/C/D are lower in energy than A’/B’/C’/D’ is right. As your values show, for PM6 calculations, A=C and B=D; the calculation method is capable of incorporating the neighbouring group effect into a hybrid structure which is treated as a non-classical carbocation. The stereoselectivity in the reaction is due to two favourable effects, A/B are both lower in energy and more reactive (better orientation and trajectory for nucleophilic attack) than A’/B’.

MP. This looks like a good mini-project – a mixture of diastereomers with rigid looking structures. The NMR analysis is good and looking at different conformations is a good idea; if the energies of these conformations have been obtained, it should even be possible to look at the Boltzmann distribution at various temperatures and work out the expected contribution to physical properties of each. Another aspect that could have been looked at here is the RMS you get when comparing calculated values for the wrong isomer to lit data; if the wrong isomer has a significantly worse correlation to the data than the correct isomer this supports the use of predicted NMR as an analytical technique in this case. The optical rotation calculations seem to be good at getting the appropriate sign of the isomer if not the absolute value; in some cases this could be used to distinguish between isomeric products.

Talk:Mod 1: Celeste van den Bosch

2011-10-27T10:42:34Z

Mjhughes:

===First week feedback===
Overall it's hard to give a good assessment of your work, mainly because there's no indication as to which answer is completed. Consequently, I'm giving you the assessment of the page 'as is' on Friday night. Some of the comments probably will be addressed by you in due course, if not already.

*The presentation of the data is good and easy to follow/mark to me. I like to see a reaction/structure scheme with numbering for each questions so I don't have to flip between pages.
*You may find that a few of your pictures could be too big, depending on the width of the browser. This is due to the fact that wiki engine doesn't scale pictures along with text. An example of a code which will help you is shown below, in which the width of your picture is 500 pixels.
<pre style="color:blue"> [[Image:filename.jpg|alignment(right/left/middle)|500px|description]] </pre>
*Question 1 - DiCy: Splitting Table 1 may help. You'll need to provide an analysis (what we really want to see!) to compare 1 with 2 and 3 with 4. Please pay attention to the source of the difference in energy, and if possible identify the part of the molecules responsible for it.
*Question 2 - Taxol: You'll need to rationalise the stability of the double bond. Pictures or Jmol to illustrate the structures in Table 2 is crucial here. MMFF94 calculation is still missing as is an explanation for the unusually stable double bond. You did notice that it's at the bridgehead and is normally strained though.
*Question 3 - MOPAC: Do we need LUMO+2? Your MO pictures looks OK, but a top-down point-of-view will probably be better. Most of your analysis is yet to be included, but what's the structure of the monoene?
*References: you've two in there. Obviously as you fill in the details and analysis, you'll need references to back your statements up.

FEEDBACK

Q1: The energy values are all correct and the discussion of selectivity in the Diels-Alder reaction is good. It is not really possible to say whether the hydrogenation reaction is under thermodynamic or kinetic control; the kinetic product (lowest energy transition state) could be the same as the thermodynamic product (lowest energy product) and in fact this is usually the case – the Diels-Alder reaction is a bit of an odd case. So if you know the product is the higher energy molecule you can say for sure that the reaction is under kinetic control since it can’t be under thermodynamic control, but if the lower energy molecule is formed you still can’t say without further analysis (transition state calculations) and more information (reaction conditions etc). As a side note: most reductions of this sort are metal-catalysed, irreversible, and therefore under kinetic control. Your comparison of the contributions to strain correctly shows the major factors (the % analysis would have been better if it had shown difference as a % of total energy – or more simply the absolute difference in kcal). The reason bending strain is worse in 3 than in 4 is because the bond angles deviate more from the sp2 ideal in that compound – due to the constraints of having a double bond in the bicyclic half of the molecule.

Q2. Your energies are all good and the favoured isomer is correctly identified. You have a brief comment on the preference for the chair conformation for the 6-ring; a more complete analysis would also show the higher energy twist-boat for compound 9; you could also have briefly described the other ways in which you attempted structural optimisation. Hyperstable alkenes are correctly defined.

Q3. The MOs look good and the IR stretch frequencies are correct. The regioselectivity in the reaction with dichlorocarbene is well explained. The difference in the C-Cl stretching frequency when the double bond is removed is due to a pi-sigma* orbital interaction. When the double bond is removed it can no longer donate into the C-Cl antibonding orbital, so the bond is stronger.

Q4. The energy values and structures all look good. R=methyl is the correct choice here for a generic alkyl group and as you have discussed semi-empirical methods are best for this case. Your finding that A/B/C/D are lower in energy than A’/B’/C’/D’ is right. As your values show, for PM6 calculations, A=C and B=D; the calculation method is capable of incorporating the neighbouring group effect into a hybrid structure which is treated as a non-classical carbocation. The stereoselectivity in the reaction is due to two favourable effects, A/B are both lower in energy and more reactive (better orientation and trajectory for nucleophilic attack) than A’/B’.

MP. This looks like a good mini-project – a mixture of diastereomers with rigid looking structures. The NMR analysis is good and looking at different conformations is a good idea; if the energies of these conformations have been obtained, it should even be possible to look at the Boltzmann distribution at various temperatures and work out the expected contribution to physical properties of each. Another aspect that could have been looked at here is the RMS you get when comparing calculated values for the wrong isomer to lit data; if the wrong isomer has a significantly worse correlation to the data than the correct isomer this supports the use of predicted NMR as an analytical technique in this case. The optical rotation calculations seem to be good at getting the appropriate sign of the isomer if not the absolute value; in some cases this could be used to distinguish between isomeric products.

Talk:Mod 1: Celeste van den Bosch

2011-10-27T10:42:20Z

Mjhughes:

===First week feedback===
Overall it's hard to give a good assessment of your work, mainly because there's no indication as to which answer is completed. Consequently, I'm giving you the assessment of the page 'as is' on Friday night. Some of the comments probably will be addressed by you in due course, if not already.

*The presentation of the data is good and easy to follow/mark to me. I like to see a reaction/structure scheme with numbering for each questions so I don't have to flip between pages.
*You may find that a few of your pictures could be too big, depending on the width of the browser. This is due to the fact that wiki engine doesn't scale pictures along with text. An example of a code which will help you is shown below, in which the width of your picture is 500 pixels.
<pre style="color:blue"> [[Image:filename.jpg|alignment(right/left/middle)|500px|description]] </pre>
*Question 1 - DiCy: Splitting Table 1 may help. You'll need to provide an analysis (what we really want to see!) to compare 1 with 2 and 3 with 4. Please pay attention to the source of the difference in energy, and if possible identify the part of the molecules responsible for it.
*Question 2 - Taxol: You'll need to rationalise the stability of the double bond. Pictures or Jmol to illustrate the structures in Table 2 is crucial here. MMFF94 calculation is still missing as is an explanation for the unusually stable double bond. You did notice that it's at the bridgehead and is normally strained though.
*Question 3 - MOPAC: Do we need LUMO+2? Your MO pictures looks OK, but a top-down point-of-view will probably be better. Most of your analysis is yet to be included, but what's the structure of the monoene?
*References: you've two in there. Obviously as you fill in the details and analysis, you'll need references to back your statements up.

FEEDBACK

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: The energy values are all correct and the discussion of selectivity in the Diels-Alder reaction is good. It is not really possible to say whether the hydrogenation reaction is under thermodynamic or kinetic control; the kinetic product (lowest energy transition state) could be the same as the thermodynamic product (lowest energy product) and in fact this is usually the case – the Diels-Alder reaction is a bit of an odd case. So if you know the product is the higher energy molecule you can say for sure that the reaction is under kinetic control since it can’t be under thermodynamic control, but if the lower energy molecule is formed you still can’t say without further analysis (transition state calculations) and more information (reaction conditions etc). As a side note: most reductions of this sort are metal-catalysed, irreversible, and therefore under kinetic control. Your comparison of the contributions to strain correctly shows the major factors (the % analysis would have been better if it had shown difference as a % of total energy – or more simply the absolute difference in kcal). The reason bending strain is worse in 3 than in 4 is because the bond angles deviate more from the sp2 ideal in that compound – due to the constraints of having a double bond in the bicyclic half of the molecule.

Q2. Your energies are all good and the favoured isomer is correctly identified. You have a brief comment on the preference for the chair conformation for the 6-ring; a more complete analysis would also show the higher energy twist-boat for compound 9; you could also have briefly described the other ways in which you attempted structural optimisation. Hyperstable alkenes are correctly defined.

Q3. The MOs look good and the IR stretch frequencies are correct. The regioselectivity in the reaction with dichlorocarbene is well explained. The difference in the C-Cl stretching frequency when the double bond is removed is due to a pi-sigma* orbital interaction. When the double bond is removed it can no longer donate into the C-Cl antibonding orbital, so the bond is stronger.

Q4. The energy values and structures all look good. R=methyl is the correct choice here for a generic alkyl group and as you have discussed semi-empirical methods are best for this case. Your finding that A/B/C/D are lower in energy than A’/B’/C’/D’ is right. As your values show, for PM6 calculations, A=C and B=D; the calculation method is capable of incorporating the neighbouring group effect into a hybrid structure which is treated as a non-classical carbocation. The stereoselectivity in the reaction is due to two favourable effects, A/B are both lower in energy and more reactive (better orientation and trajectory for nucleophilic attack) than A’/B’.

MP. This looks like a good mini-project – a mixture of diastereomers with rigid looking structures. The NMR analysis is good and looking at different conformations is a good idea; if the energies of these conformations have been obtained, it should even be possible to look at the Boltzmann distribution at various temperatures and work out the expected contribution to physical properties of each. Another aspect that could have been looked at here is the RMS you get when comparing calculated values for the wrong isomer to lit data; if the wrong isomer has a significantly worse correlation to the data than the correct isomer this supports the use of predicted NMR as an analytical technique in this case. The optical rotation calculations seem to be good at getting the appropriate sign of the isomer if not the absolute value; in some cases this could be used to distinguish between isomeric products.

Talk:Mod:mw1409module1

2011-10-27T10:40:04Z

Mjhughes:

FEEDBACK

Q1: Your energy values are all good, but you could have shown at least an image of the structures. The dimerisation is indeed under kinetic control; this is specifically due to stabilising secondary orbital interactions in the Diels-Alder transition state that leads to the endo-product. Bending strain is identified as the key difference between compounds 3 and 4. The reason for this is in fact the double bond in the bridged bicycle deviates further from its ideal sp2 bond angles than the one in the fused 5-ring does.

Q2. The energies are fine and the lowest energy isomer is correctly predicted. You said that other conformations didn’t result in lower energies, but it would have helped to include these results to give a more complete answer that showed your approach to looking for different conformations. Although in this case the lowest energies are obtained with chair conformations for the 6-ring, this is not necessarily true in all cases; some substituents and substitution patterns can lead to the twist-boat or even the boat form being favoured. The definition of a hyperstable alkene is correct.

Q3. The MOs and IR stretching energies are good. Getting symmetrical MOs can be problematic with the methods employed here, but the key thing was to recognise that they weren’t correct! The explanation for the regiocontrol in the reaction with dichlorocarbene is right. The reason that the C-Cl stretch becomes stronger on removal of the anti double bond is that there is an interaction between that pi bond and the C-Cl sigma* which weakens the bond.

Q4. Your energies and structures all look good. I would have recommended that R=methyl is the best choice for a generic methyl group, but your analysis shows that the same results are observed with R=hydrogen. As you found, A/B/C/D have lower energies than A’/B’/C’/D’ and for C’ and D’ the trans-fused bicycles are significantly more energetic. As your numbers and structures indicate when you use PM6 to model these intermediates you find that A=C and B=D. This is because, the semi-empirical method accounts for the neighbouring group effect and the structure calculated is a non-classical carbocation. The origin of the stereoselectivity for the formation of C and D is described in your text but not explicitly referred to as such! The intermediates A and B are both lower in energy and also more reactive (better orientation and angle for nucleophilic attack) than A’ and B’.

MP. You discussed this interesting difference in stereochemical outcome depending on which reducing agent is employed and it would have been nice to see some more discussion of this and how computational chemistry could be employed to elucidate this aspect. Your NMR analysis would have benefitted from a more quantitative approach. Comparison of the lit and calculated values would be best shown as the differences in ppm which could be tabulated, or better displayed graphically. Additionally, in order to show that this method could be used to distinguish between the two isomers it is necessary to show that the calculated data matches the actual data for the isomer better than calculated data for the wrong isomer.

Talk:Mod:mw1409module1

2011-10-27T10:39:34Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your energy values are all good, but you could have shown at least an image of the structures. The..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your energy values are all good, but you could have shown at least an image of the structures. The dimerisation is indeed under kinetic control; this is specifically due to stabilising secondary orbital interactions in the Diels-Alder transition state that leads to the endo-product. Bending strain is identified as the key difference between compounds 3 and 4. The reason for this is in fact the double bond in the bridged bicycle deviates further from its ideal sp2 bond angles than the one in the fused 5-ring does.
Q2. The energies are fine and the lowest energy isomer is correctly predicted. You said that other conformations didn’t result in lower energies, but it would have helped to include these results to give a more complete answer that showed your approach to looking for different conformations. Although in this case the lowest energies are obtained with chair conformations for the 6-ring, this is not necessarily true in all cases; some substituents and substitution patterns can lead to the twist-boat or even the boat form being favoured. The definition of a hyperstable alkene is correct.
Q3. The MOs and IR stretching energies are good. Getting symmetrical MOs can be problematic with the methods employed here, but the key thing was to recognise that they weren’t correct! The explanation for the regiocontrol in the reaction with dichlorocarbene is right. The reason that the C-Cl stretch becomes stronger on removal of the anti double bond is that there is an interaction between that pi bond and the C-Cl sigma* which weakens the bond.
Q4. Your energies and structures all look good. I would have recommended that R=methyl is the best choice for a generic methyl group, but your analysis shows that the same results are observed with R=hydrogen. As you found, A/B/C/D have lower energies than A’/B’/C’/D’ and for C’ and D’ the trans-fused bicycles are significantly more energetic. As your numbers and structures indicate when you use PM6 to model these intermediates you find that A=C and B=D. This is because, the semi-empirical method accounts for the neighbouring group effect and the structure calculated is a non-classical carbocation. The origin of the stereoselectivity for the formation of C and D is described in your text but not explicitly referred to as such! The intermediates A and B are both lower in energy and also more reactive (better orientation and angle for nucleophilic attack) than A’ and B’.
MP. You discussed this interesting difference in stereochemical outcome depending on which reducing agent is employed and it would have been nice to see some more discussion of this and how computational chemistry could be employed to elucidate this aspect. Your NMR analysis would have benefitted from a more quantitative approach. Comparison of the lit and calculated values would be best shown as the differences in ppm which could be tabulated, or better displayed graphically. Additionally, in order to show that this method could be used to distinguish between the two isomers it is necessary to show that the calculated data matches the actual data for the isomer better than calculated data for the wrong isomer.

Talk:Mod:YYTmod1

2011-10-27T10:38:41Z

Mjhughes:

FEEDBACK

Q1: Your calculated energies are good, but it is not necessary to give so many significant figures; MM2 is an approximate method and there is some degree of error to consider. You should have included at least a jpg image of your structures and ideally a jmol or link to a jmol structure. The differences between the isomers are discussed, but it would have been better to relate this analysis to the structures of the compounds in question. The discussion of kinetic vs thermodynamic control is good for the dimerisation reaction. For the hydrogenation, it is not really possible to comment without further information about the conditions. Normally this type of reduction is metal-catalysed, irreversible, and therefore under kinetic control. To predict the kinetic product, DFT methods would need to be used to calculate the relevant transition state energies.

Q2. The lowest energy isomer is correctly identified, but the structures are not correct. The lowest energy forms of both isomers have the 6-ring in a chair conformation not a twist-boat conformation. There is no discussion here of how you optimised the compounds – since there is not much more to discuss in this question this would have added a bit more detail. The reason the oxy-cope rearrangement you have shown is often irreversible is that the product (an enol) rapidly isomerises to a more stable form (the appropriate carbonyl compound) which is unreactive in the backwards sense. In this case, the double bond is unreactive in a more general sense (to any reagents); this is because it is a “hyperstable alkene” – an alkene for which the parent alkane is more strained due to the bridgehead position on a medium –sized ring.

Q3. Your calculations, MOs and IR stretches are all good. The explanation for the regioselectivity in the reaction with dichlorocarbene is correct. You have identified the pi-sigma* interaction that lowers the C-Cl bond energy, but did not correlate this with the IR data. The IR stretch of the C-Cl bond is more energetic (stronger bond) when the double bond is removed because the donation into its antibonding orbital is removed.

Q4. The choice of R=methyl is correct and the PM6 calculations are indeed more appropriate than MM2. As you have shown the A/B/C/D set is lower in energy than the A*/B*/C*/D* set. Your values are overall higher than expected; I suspect that the carbonyl oxygen of the acetyl group is not sufficiently close to the oxonium carbon. The structures C* and D* were the trans fused bicycles – any fused bicycle can potentially be cis or trans-fused for 6.5 structures, the trans-fused scenario is usually much higher in energy. The table of bond distances is not helpful – it would be better to list the relevant properties you want to discuss or at least highlight them in the text. Comparison of data to ideal bond lengths and angles becomes less relevant when using semi-empirical methods as you are no longer relying solely on a set of empirical reference points.

MP. This is a good simple reaction for a mini-project as it produces a mixture of isomers which are conformationally restricted. However, I’m not sure how well simple computational methods can account for an ion pair. You are right in saying that the diastereomers could be distinguished by different coupling constants. Looking at your NMR spectra, I think it is possible that you didn’t apply the TMS reference which will dramatically change the ppm values. When comparing lit and calculated NMR data it is better to give the differences in ppm rather than just both sets of data; the clearest way to do this is to present the differences graphically, so that the error and the overall deviation can be easily assessed.

Talk:Mod:YYTmod1

2011-10-27T10:38:16Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your calculated energies are good, but it is not necessary to give so many significant figures; MM2..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your calculated energies are good, but it is not necessary to give so many significant figures; MM2 is an approximate method and there is some degree of error to consider. You should have included at least a jpg image of your structures and ideally a jmol or link to a jmol structure. The differences between the isomers are discussed, but it would have been better to relate this analysis to the structures of the compounds in question. The discussion of kinetic vs thermodynamic control is good for the dimerisation reaction. For the hydrogenation, it is not really possible to comment without further information about the conditions. Normally this type of reduction is metal-catalysed, irreversible, and therefore under kinetic control. To predict the kinetic product, DFT methods would need to be used to calculate the relevant transition state energies.

Q2. The lowest energy isomer is correctly identified, but the structures are not correct. The lowest energy forms of both isomers have the 6-ring in a chair conformation not a twist-boat conformation. There is no discussion here of how you optimised the compounds – since there is not much more to discuss in this question this would have added a bit more detail. The reason the oxy-cope rearrangement you have shown is often irreversible is that the product (an enol) rapidly isomerises to a more stable form (the appropriate carbonyl compound) which is unreactive in the backwards sense. In this case, the double bond is unreactive in a more general sense (to any reagents); this is because it is a “hyperstable alkene” – an alkene for which the parent alkane is more strained due to the bridgehead position on a medium –sized ring.

Q3. Your calculations, MOs and IR stretches are all good. The explanation for the regioselectivity in the reaction with dichlorocarbene is correct. You have identified the pi-sigma* interaction that lowers the C-Cl bond energy, but did not correlate this with the IR data. The IR stretch of the C-Cl bond is more energetic (stronger bond) when the double bond is removed because the donation into its antibonding orbital is removed.

Q4. The choice of R=methyl is correct and the PM6 calculations are indeed more appropriate than MM2. As you have shown the A/B/C/D set is lower in energy than the A*/B*/C*/D* set. Your values are overall higher than expected; I suspect that the carbonyl oxygen of the acetyl group is not sufficiently close to the oxonium carbon. The structures C* and D* were the trans fused bicycles – any fused bicycle can potentially be cis or trans-fused for 6.5 structures, the trans-fused scenario is usually much higher in energy. The table of bond distances is not helpful – it would be better to list the relevant properties you want to discuss or at least highlight them in the text. Comparison of data to ideal bond lengths and angles becomes less relevant when using semi-empirical methods as you are no longer relying solely on a set of empirical reference points.

MP. This is a good simple reaction for a mini-project as it produces a mixture of isomers which are conformationally restricted. However, I’m not sure how well simple computational methods can account for an ion pair. You are right in saying that the diastereomers could be distinguished by different coupling constants. Looking at your NMR spectra, I think it is possible that you didn’t apply the TMS reference which will dramatically change the ppm values. When comparing lit and calculated NMR data it is better to give the differences in ppm rather than just both sets of data; the clearest way to do this is to present the differences graphically, so that the error and the overall deviation can be easily assessed.

Talk:Mod:YJY1109

2011-10-27T10:37:25Z

Mjhughes:

FEEDBACK


Q1: Your values are all good and the factors that contribute to strain and differentiate the different isomers are correctly identified and explained. The discussion of kinetic vs thermodynamic control is correct, although it is not really possible to comment much on the selectivity of the hydrogenation reaction without further information. If the reaction is under thermodynamic control the lowest energy isomer you show would be the predicted product. Hydrogenations of this type are typically, metal catalysed, irreversible, and therefore under kinetic control. To predict the outcome of this type of reaction, the transition states would need to be calculated using a DFT method in order to find the kinetically favoured product.

Q2. The lowest energy isomer is correctly identified and the energy values are good. Although, the chair conformation is the lowest energy form in both cases here, this is not necessarily true for all 6-rings. In some cases, the specific substituents can mean that the twist-boat or even the boat form is the most stable conformation. The analysis of 6-ring conformation is a rational starting point for this energy minimisation, but it would have been nice to hear a little more about other ways in which you attempted to get the lowest energy form. Hyperstable olefins are correctly defined and it would be possible to get a value for the olefin strain if you had performed DFT calculations. However, with the molecular mechanics calculations you have it is not possible, because strictly you can only compare isomers.

Q3. This was very well answered. The MOs and IR stretches are good, the rationalisation of the reactivity with dichlorocarbene is correct, and the analysis of the IR stretch changes with and without the pi-sigma* interaction is correct.

Q4. Your calculated structures and energies are a fair bit off. It is possible to twist the acetyl group so that its carbonyl oxygen points above or below the oxonium carbon in all cases without employing a ring-flip of the other substituents. Also, for C* and D*, the idea was to perform calculations on the trans-fused [6.5] bicyclic system in comparison to the cis-fused [6.5] bicycles C and D. Do not hesitate to ask a demonstrator if there is any similar confusion about questions in future modules. R=methyl and semi-empirical calculations were correct choices for this problem for the reasons you gave. Diastereoselectivity in the formation of C/D over C*/D* is due to the intermediates A/B being favoured thermodynamically and also being more reactive (better trajectory and distance for nucleophilic attack).

MP. Although you have listed the calculated and lit NMR values, it would be better to show the differences in ppm. This can be done graphically to good effect, showing instantly the error and the overall deviation. In order to assess the use of NMR prediction to distinguish between different isomers, it would be necessary to compare the calculated data of both isomers to the lit data. If the prediction technique is valid, the error in the comparison using the actual isomer should be less than in the comparison of the false isomer.

Talk:Mod:YJY1109

2011-10-27T10:37:05Z

Mjhughes:

FEEDBACK

Q1: Your values are all good and the factors that contribute to strain and differentiate the different isomers are correctly identified and explained. The discussion of kinetic vs thermodynamic control is correct, although it is not really possible to comment much on the selectivity of the hydrogenation reaction without further information. If the reaction is under thermodynamic control the lowest energy isomer you show would be the predicted product. Hydrogenations of this type are typically, metal catalysed, irreversible, and therefore under kinetic control. To predict the outcome of this type of reaction, the transition states would need to be calculated using a DFT method in order to find the kinetically favoured product.

Q2. The lowest energy isomer is correctly identified and the energy values are good. Although, the chair conformation is the lowest energy form in both cases here, this is not necessarily true for all 6-rings. In some cases, the specific substituents can mean that the twist-boat or even the boat form is the most stable conformation. The analysis of 6-ring conformation is a rational starting point for this energy minimisation, but it would have been nice to hear a little more about other ways in which you attempted to get the lowest energy form. Hyperstable olefins are correctly defined and it would be possible to get a value for the olefin strain if you had performed DFT calculations. However, with the molecular mechanics calculations you have it is not possible, because strictly you can only compare isomers.

Q3. This was very well answered. The MOs and IR stretches are good, the rationalisation of the reactivity with dichlorocarbene is correct, and the analysis of the IR stretch changes with and without the pi-sigma* interaction is correct.

Q4. Your calculated structures and energies are a fair bit off. It is possible to twist the acetyl group so that its carbonyl oxygen points above or below the oxonium carbon in all cases without employing a ring-flip of the other substituents. Also, for C* and D*, the idea was to perform calculations on the trans-fused [6.5] bicyclic system in comparison to the cis-fused [6.5] bicycles C and D. Do not hesitate to ask a demonstrator if there is any similar confusion about questions in future modules. R=methyl and semi-empirical calculations were correct choices for this problem for the reasons you gave. Diastereoselectivity in the formation of C/D over C*/D* is due to the intermediates A/B being favoured thermodynamically and also being more reactive (better trajectory and distance for nucleophilic attack).

MP. Although you have listed the calculated and lit NMR values, it would be better to show the differences in ppm. This can be done graphically to good effect, showing instantly the error and the overall deviation. In order to assess the use of NMR prediction to distinguish between different isomers, it would be necessary to compare the calculated data of both isomers to the lit data. If the prediction technique is valid, the error in the comparison using the actual isomer should be less than in the comparison of the false isomer.

Talk:Mod:YJY1109

2011-10-27T10:36:49Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your values are all good and the factors that contribute to strain and differentiate the different ..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your values are all good and the factors that contribute to strain and differentiate the different isomers are correctly identified and explained. The discussion of kinetic vs thermodynamic control is correct, although it is not really possible to comment much on the selectivity of the hydrogenation reaction without further information. If the reaction is under thermodynamic control the lowest energy isomer you show would be the predicted product. Hydrogenations of this type are typically, metal catalysed, irreversible, and therefore under kinetic control. To predict the outcome of this type of reaction, the transition states would need to be calculated using a DFT method in order to find the kinetically favoured product.

Q2. The lowest energy isomer is correctly identified and the energy values are good. Although, the chair conformation is the lowest energy form in both cases here, this is not necessarily true for all 6-rings. In some cases, the specific substituents can mean that the twist-boat or even the boat form is the most stable conformation. The analysis of 6-ring conformation is a rational starting point for this energy minimisation, but it would have been nice to hear a little more about other ways in which you attempted to get the lowest energy form. Hyperstable olefins are correctly defined and it would be possible to get a value for the olefin strain if you had performed DFT calculations. However, with the molecular mechanics calculations you have it is not possible, because strictly you can only compare isomers.

Q3. This was very well answered. The MOs and IR stretches are good, the rationalisation of the reactivity with dichlorocarbene is correct, and the analysis of the IR stretch changes with and without the pi-sigma* interaction is correct.

Q4. Your calculated structures and energies are a fair bit off. It is possible to twist the acetyl group so that its carbonyl oxygen points above or below the oxonium carbon in all cases without employing a ring-flip of the other substituents. Also, for C* and D*, the idea was to perform calculations on the trans-fused [6.5] bicyclic system in comparison to the cis-fused [6.5] bicycles C and D. Do not hesitate to ask a demonstrator if there is any similar confusion about questions in future modules. R=methyl and semi-empirical calculations were correct choices for this problem for the reasons you gave. Diastereoselectivity in the formation of C/D over C*/D* is due to the intermediates A/B being favoured thermodynamically and also being more reactive (better trajectory and distance for nucleophilic attack).

MP. Although you have listed the calculated and lit NMR values, it would be better to show the differences in ppm. This can be done graphically to good effect, showing instantly the error and the overall deviation. In order to assess the use of NMR prediction to distinguish between different isomers, it would be necessary to compare the calculated data of both isomers to the lit data. If the prediction technique is valid, the error in the comparison using the actual isomer should be less than in the comparison of the false isomer.

Talk:Mod:CSWmodule1

2011-10-27T10:36:09Z

Mjhughes:

FEEDBACK

Q1: Your energy values are all spot on and the discussion of contributions to strain is good. The discussion of kinetic vs thermodynamic control is all correct and as you point out, it is not possible to comment further on the outcome of the hydrogenation reaction without more information. The majority of such reactions are irreversible and therefore under kinetic control. It would therefore be necessary to calculate transition state energies to find out which product is preferred.

Q2. Your energies are good as is the discussion of your optimisation approach – this was the best way to answer this question with brief description of your calculations. Analysis of the energy contributions as in Q1 would have been a nice addition here. Hyperstable alkenes are well defined.

Q3. The MOs and IR stretch values are good – you should have included the energies of the main compound obtained with different calculation methods; this isn’t very important in this question, but it is a way in which the calculations are assessed and you may lose marks in the future modules with an omission of this type. The different double bond reactivities is discussed well. The IR stretches are indeed effected by an interaction with the C-Cl sigma* orbital: The pi orbital of the anti double bond is suitably positioned to interact with this antibonding orbital and therefore the C-Cl bond is weaker, and the stretch energy is lower when there is a double bond in that position.

Q4. The combination of R=methyl and semi-empirical methods is indeed the best choice here. Your structures are fine, but better approximations to the problem are found when the carbonyl oxygen of the neighbouring acetyl group is placed in a position where it can attack the oxonium carbon. The jmols for C/C*/D/D* are actually structures for A/A*/B/B*. The energy values you got are nevertheless reasonably close to those expected and you correctly found that the set A/B/C/D is lower in energy than A*/B*/C*/D*. The selectivity in this process is due to a combination of lower energy and better reactivity of the intermediates A and B compared to A* and B*. Using PM6 you should have found that A=C and B=D because the structure calculated incorporates the neighbouring group participation and treats the intermediate as a non-classical carbocation.

MP. You have compared lit and calculated data for both isomers and shown that there is a moderate correlation. It would be good to see whether the calculated spectrum for one isomer is more closely matched to the isomer it is supposed to be than to the other isomer. This would be the key to having an accurate method for assigning the stereochemistry of product of unknown configuration. It is good that you have shown the differences in ppm but there are lots of ways to analyse the differences more quantitatively to give a more general figure to describe the accuracy of the calculation. Something else to consider is how you would distinguish between these isomers experimentally.

Talk:Mod:CSWmodule1

2011-10-27T10:35:54Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE ..."

Normal 0 false false false EN-GB X-NONE X-NONE 
FEEDBACK

Q1: Your energy values are all spot on and the discussion of contributions to strain is good. The discussion of kinetic vs thermodynamic control is all correct and as you point out, it is not possible to comment further on the outcome of the hydrogenation reaction without more information. The majority of such reactions are irreversible and therefore under kinetic control. It would therefore be necessary to calculate transition state energies to find out which product is preferred.

Q2. Your energies are good as is the discussion of your optimisation approach – this was the best way to answer this question with brief description of your calculations. Analysis of the energy contributions as in Q1 would have been a nice addition here. Hyperstable alkenes are well defined.

Q3. The MOs and IR stretch values are good – you should have included the energies of the main compound obtained with different calculation methods; this isn’t very important in this question, but it is a way in which the calculations are assessed and you may lose marks in the future modules with an omission of this type. The different double bond reactivities is discussed well. The IR stretches are indeed effected by an interaction with the C-Cl sigma* orbital: The pi orbital of the anti double bond is suitably positioned to interact with this antibonding orbital and therefore the C-Cl bond is weaker, and the stretch energy is lower when there is a double bond in that position.

Q4. The combination of R=methyl and semi-empirical methods is indeed the best choice here. Your structures are fine, but better approximations to the problem are found when the carbonyl oxygen of the neighbouring acetyl group is placed in a position where it can attack the oxonium carbon. The jmols for C/C*/D/D* are actually structures for A/A*/B/B*. The energy values you got are nevertheless reasonably close to those expected and you correctly found that the set A/B/C/D is lower in energy than A*/B*/C*/D*. The selectivity in this process is due to a combination of lower energy and better reactivity of the intermediates A and B compared to A* and B*. Using PM6 you should have found that A=C and B=D because the structure calculated incorporates the neighbouring group participation and treats the intermediate as a non-classical carbocation.

MP. You have compared lit and calculated data for both isomers and shown that there is a moderate correlation. It would be good to see whether the calculated spectrum for one isomer is more closely matched to the isomer it is supposed to be than to the other isomer. This would be the key to having an accurate method for assigning the stereochemistry of product of unknown configuration. It is good that you have shown the differences in ppm but there are lots of ways to analyse the differences more quantitatively to give a more general figure to describe the accuracy of the calculation. Something else to consider is how you would distinguish between these isomers experimentally.

Talk:Mod:gjsmodule1

2011-10-27T10:34:49Z

Mjhughes:

FEEDBACK

Q1: Your energy values are all correct although there is no need to report so many significant figures – there is some error to account for given that this is an approximation and the values won’t be the same beyond a certain point if the calculation is repeated from scratch. The exo- and endo- dimers are not conformational isomers – they are diastereoisomers. The main difference between the strain between the monohydrogenated compounds is correctly identified as bending-strain and well explained. The discussion of kinetic vs thermodynamic control is spot on: you correctly point out that it is not possible to comment on the outcome of the hydrogenation without more information. Most hydrogenation conditions involve a metal catalyst and gaseous hydrogen, are irreversible, and therefore under kinetic control. Modelling with DFT methods would allow calculation of transition state energies and therefore prediction of the kinetic product.

Q2. Your calculated values for compound 10 are good but the lowest energy structure for 9 is too low; in fact 9 should be the higher energy isomer – as you found with the different molecular mechanics force field. It would have been beneficial to see some jmol structures to see why this is the case. 6-ring conformations are a logical focus for optimisation because there are known preferences to analyse – some addition discussion of other ways in which you optimised the compounds would have been good. Hyperstable alkenes are correctly defined.

Q3. The calculations, MOs and IR stretches look good. You should make sure you report all of the energies you calculate in future modules – as these numbers are one of the ways the reports are assessed and without other results as in this question it would be difficult to work out whether the calculations are correct. The discussions of double bond nucleophilicity and the pi-sigma* interaction are good and the IR stretch differences are rationalised well.

Q4. Your assessment of A/B/C/D vs A*/B*/C*/D* is correct with the former group all being lower in energy. The energies you got are a little high, there are lower energy conformations accessible to some of your isomers, in particular when the carbonyl oxygen of the acetyl group is moved closer to the oxonium carbon. Also some of your * structures are not conformations in which the orientation of the acetyl carbonyl oxygen has switched. The combination of R=methyl and semi-empirical calculations is the correct choice for this question. You should have found that using PM6, A=C and B=D; the neighbouring group effect is directly incorporated into the structure which is a non-classical carbocation. Selectivity in the reaction is due to a combination of favoured formation of A/B/C/D over A*/B*/C*/D* and also higher reactivity of A and B due to more favourable orientation for attack on the oxonium group. This is reflected in your finding that C* and D* are much higher in energy, precluding reaction via these pathways.

MP. The comparison of the lit data for isomer A with the calculated data for A, B, C, and D is an interesting question for a mini-project and is exactly the type of scenario where this kind of NMR calculation is applied in computational research. It seems that it is difficult to distinguish these isomers by calculating NMR data – if the actual data are very close it will be impossible no matter how accurate the methods and calculations. Your analysis would have benefited from some kind of analysis of the error between the calculated and lit values. A graphical representation of the ppm differences is often the best way to do this, but more sophisticated statistical analyses are possible. As you have found, the calculated IR data does not seem to be accurate enough for this kind of investigation. The differences in energy you found for these isomers are not insignificant: In conformational analysis, a difference in energy of around 2 kcal/mol will correspond to a >9:1 preference for the lower energy for the lower energy isomer at room temperature. It is typical for comparison of energies to set one isomer at 0 and then give the other values relative to this standard.

Talk:Mod:gjsmodule1

2011-10-27T10:34:23Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your energy values are all correct although there is no need to report so many significant figures ..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your energy values are all correct although there is no need to report so many significant figures – there is some error to account for given that this is an approximation and the values won’t be the same beyond a certain point if the calculation is repeated from scratch. The exo- and endo- dimers are not conformational isomers – they are diastereoisomers. The main difference between the strain between the monohydrogenated compounds is correctly identified as bending-strain and well explained. The discussion of kinetic vs thermodynamic control is spot on: you correctly point out that it is not possible to comment on the outcome of the hydrogenation without more information. Most hydrogenation conditions involve a metal catalyst and gaseous hydrogen, are irreversible, and therefore under kinetic control. Modelling with DFT methods would allow calculation of transition state energies and therefore prediction of the kinetic product.

Q2. Your calculated values for compound 10 are good but the lowest energy structure for 9 is too low; in fact 9 should be the higher energy isomer – as you found with the different molecular mechanics force field. It would have been beneficial to see some jmol structures to see why this is the case. 6-ring conformations are a logical focus for optimisation because there are known preferences to analyse – some addition discussion of other ways in which you optimised the compounds would have been good. Hyperstable alkenes are correctly defined.

Q3. The calculations, MOs and IR stretches look good. You should make sure you report all of the energies you calculate in future modules – as these numbers are one of the ways the reports are assessed and without other results as in this question it would be difficult to work out whether the calculations are correct. The discussions of double bond nucleophilicity and the pi-sigma* interaction are good and the IR stretch differences are rationalised well.

Q4. Your assessment of A/B/C/D vs A*/B*/C*/D* is correct with the former group all being lower in energy. The energies you got are a little high, there are lower energy conformations accessible to some of your isomers, in particular when the carbonyl oxygen of the acetyl group is moved closer to the oxonium carbon. Also some of your * structures are not conformations in which the orientation of the acetyl carbonyl oxygen has switched. The combination of R=methyl and semi-empirical calculations is the correct choice for this question. You should have found that using PM6, A=C and B=D; the neighbouring group effect is directly incorporated into the structure which is a non-classical carbocation. Selectivity in the reaction is due to a combination of favoured formation of A/B/C/D over A*/B*/C*/D* and also higher reactivity of A and B due to more favourable orientation for attack on the oxonium group. This is reflected in your finding that C* and D* are much higher in energy, precluding reaction via these pathways.

MP. The comparison of the lit data for isomer A with the calculated data for A, B, C, and D is an interesting question for a mini-project and is exactly the type of scenario where this kind of NMR calculation is applied in computational research. It seems that it is difficult to distinguish these isomers by calculating NMR data – if the actual data are very close it will be impossible no matter how accurate the methods and calculations. Your analysis would have benefited from some kind of analysis of the error between the calculated and lit values. A graphical representation of the ppm differences is often the best way to do this, but more sophisticated statistical analyses are possible. As you have found, the calculated IR data does not seem to be accurate enough for this kind of investigation. The differences in energy you found for these isomers are not insignificant: In conformational analysis, a difference in energy of around 2 kcal/mol will correspond to a >9:1 preference for the lower energy for the lower energy isomer at room temperature. It is typical for comparison of energies to set one isomer at 0 and then give the other values relative to this standard.

Talk:Mod:3516464

2011-10-27T10:33:19Z

Mjhughes:

FEEDBACK

Good introduction – the differences between molecular mechanics and DFT calculations are clearly stated.

Q1: Your calculations and structures are all good. The discussion of kinetic vs thermodynamic control is good although the following statement is not necessarily true: “Therefore it can be rationalized that compound 4 is the more stable dominant product, and the hydrogenation reaction is under thermodynamic control.” The hydrogenation reaction could still be under kinetic control and it could be that the thermodynamic product is the same as the kinetic product. This is in fact often the case and the D-A reaction is a bit of an oddity. So it is not possible to definitively say the reaction is under thermodynamic control or kinetic control without carrying out further calculations on the relevant transition states (not possible for MM2). Your analysis of the strain-contributors in the fully hydrogenated compound is valid, but it is not possible to compare the energy of this product to the monohydrogenated compounds because they are not isomeric.

Q2. Your energies are good and the “down” isomer is correctly identified as the lowest energy one. In fact your lowest energy for compound 9 is lower than expected! The atropisomerism has nothing to do with the hyperstability of the olefin; this type of isomerisim merely arises when there are conformational isomers which “don’t interconvert”. The non-interconversion is somewhat arbitrary and time-dependent: As you have said the isomers do interconvert, but only after a certain time period (usually of course, conformational isomers interconvert rapidly). The 6-ring is a good focus for structural optimisation because there are known conformations to look for; it would have been good to hear a brief description of other ways in which you optimised the structure. The definition of alkene hyperstability is right. The calculated energy of the hydrogenated form can’t strictly be compared to the energy of the alkene because you can only compare isomers with molecular mechanics. Structural analysis is still a valid method however and comparison of sp2 and sp3 bond deviations is a good approach.

Q3. Your calculations, MOs, and IR stretch values look good. The discussion of the pi-sigma* interaction is good and its impact on the vibrational energies is detailed well. NB: Anti-periplanar is a specific term which refers to a 180 deg dihedral angle between two groups; it should only be used when talking about atoms that are separated by 3 consecutive bonds.

Q4. The energy values you got are close to those expected and as you have found the A/B/C/D set is expected to be lower in energy than A’/B’/C’/D’. Methyl is the best choice of alkyl group for the reasons you stated and the semi-empirical method is more appropriate than molecular mechanics here. As you can see from your energy values and structures with MOPAC, A=C and B=D; the neighbouring group interaction is directly incorporated into a hybrid structure because MOPAC treats the system (more realistically) as a non-classical carbocation. The reason for the stereoselectivity is partly that the intermediate conformations A and B are the lowest in energy (compared to A’ and B’) and also that these intermediates have a better orientation for nucleophilic attack; hence these intermediates are favoured on both sides of the Curtin-Hammett kinetics dichotomy.

MP. This kind of attempt to distinguish diastereoisomers by calculating physical properties is exactly the right approach for a mini-project. The NMR data seems to be a good match to both isomers as indicated by your graph. Perhaps a clearer case could be made by graphically comparing the absolute differences from the calculated and lit values to see which isomer differs the most and whether it is a significant difference. On that point there are many error analyses which would give a more quantitative answer. The IR spectra, while worth including were unlikely to help you to distinguish between the isomers, but it is interesting that your optical rotation give a strong indication as to which isomer is formed. For your discussion of the diastereoselectivity, it is always useful to refer to a detailed reaction mechanism to help your analysis. Another point that would have been worth discussing is possible ways to tell these isomers apart experimentally and whether this was done by the authors of the synthesis.

Talk:Mod:3516464

2011-10-27T10:32:57Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Good introduction – the differences between molecular mechanics and DFT calculations are clearly stat..."

Normal 0 false false false EN-GB X-NONE X-NONE

Good introduction – the differences between molecular mechanics and DFT calculations are clearly stated.

Q1: Your calculations and structures are all good. The discussion of kinetic vs thermodynamic control is good although the following statement is not necessarily true: “Therefore it can be rationalized that compound 4 is the more stable dominant product, and the hydrogenation reaction is under thermodynamic control.” The hydrogenation reaction could still be under kinetic control and it could be that the thermodynamic product is the same as the kinetic product. This is in fact often the case and the D-A reaction is a bit of an oddity. So it is not possible to definitively say the reaction is under thermodynamic control or kinetic control without carrying out further calculations on the relevant transition states (not possible for MM2). Your analysis of the strain-contributors in the fully hydrogenated compound is valid, but it is not possible to compare the energy of this product to the monohydrogenated compounds because they are not isomeric.

Q2. Your energies are good and the “down” isomer is correctly identified as the lowest energy one. In fact your lowest energy for compound 9 is lower than expected! The atropisomerism has nothing to do with the hyperstability of the olefin; this type of isomerisim merely arises when there are conformational isomers which “don’t interconvert”. The non-interconversion is somewhat arbitrary and time-dependent: As you have said the isomers do interconvert, but only after a certain time period (usually of course, conformational isomers interconvert rapidly). The 6-ring is a good focus for structural optimisation because there are known conformations to look for; it would have been good to hear a brief description of other ways in which you optimised the structure. The definition of alkene hyperstability is right. The calculated energy of the hydrogenated form can’t strictly be compared to the energy of the alkene because you can only compare isomers with molecular mechanics. Structural analysis is still a valid method however and comparison of sp2 and sp3 bond deviations is a good approach.

Q3. Your calculations, MOs, and IR stretch values look good. The discussion of the pi-sigma* interaction is good and its impact on the vibrational energies is detailed well. NB: Anti-periplanar is a specific term which refers to a 180 deg dihedral angle between two groups; it should only be used when talking about atoms that are separated by 3 consecutive bonds.

Q4. The energy values you got are close to those expected and as you have found the A/B/C/D set is expected to be lower in energy than A’/B’/C’/D’. Methyl is the best choice of alkyl group for the reasons you stated and the semi-empirical method is more appropriate than molecular mechanics here. As you can see from your energy values and structures with MOPAC, A=C and B=D; the neighbouring group interaction is directly incorporated into a hybrid structure because MOPAC treats the system (more realistically) as a non-classical carbocation. The reason for the stereoselectivity is partly that the intermediate conformations A and B are the lowest in energy (compared to A’ and B’) and also that these intermediates have a better orientation for nucleophilic attack; hence these intermediates are favoured on both sides of the Curtin-Hammett kinetics dichotomy.

MP. This kind of attempt to distinguish diastereoisomers by calculating physical properties is exactly the right approach for a mini-project. The NMR data seems to be a good match to both isomers as indicated by your graph. Perhaps a clearer case could be made by graphically comparing the absolute differences from the calculated and lit values to see which isomer differs the most and whether it is a significant difference. On that point there are many error analyses which would give a more quantitative answer. The IR spectra, while worth including were unlikely to help you to distinguish between the isomers, but it is interesting that your optical rotation give a strong indication as to which isomer is formed. For your discussion of the diastereoselectivity, it is always useful to refer to a detailed reaction mechanism to help your analysis. Another point that would have been worth discussing is possible ways to tell these isomers apart experimentally and whether this was done by the authors of the synthesis.

Talk:Mod:HAS1501

2011-10-27T10:31:40Z

Mjhughes:

FEEDBACK

Q1: Your calculated values are all good, although there is no need to include so many significant figures: There is some error in the calculation which is after all an approximation and you have probably seen using ChemBio3D that the latter decimal figures won’t be the same if you do the same calculation twice from scratch. The major difference between compounds 3 and 4 is correctly identified as bending strain – for a more complete answer, the structures could be compared to find the origin of this difference, namely the difference bond angles in the double bonds present. Your comment on the hydrogenation reaction is right – under thermodynamic control you would get 4 and in order to get compound 3 the reaction would have to be under kinetic control.

Q2. Your calculated structures are not the lowest energy forms although isomer B is indeed the lowest energy one. Both of your compounds, in fact, have the 6-ring in the twist-boat form. In the lowest energy form the 6-rings should be in the chair conformation. It would have been worth detailing briefly the measures you took to minimise the structures, an analysis of the different possibilities for the 6-ring would have lead you to the right answer. The discussion of the strain contributions is good. Hyperstable alkenes are well defined, but it is not possible to carry out that comparison of the hydrogenated “parent” compounds using molecular mechanics, because the compounds are not isomeric. It would be possible using the absolute scale of a DFT calculation.

Q3. Your MOs look good, as do the calculated IR spectra. The energies obtained for the main compound with the different calculation methods are not reported – in the future modules make sure you report these different energy values if appropriate, as you may be assessed on the numbers. You have identified the possibility of a pi-sigma* interaction, but did not discuss the impact this has on your various IR stretches. The anti pi bond has a lower energy stretch than the syn pi bond and the C-Cl stretch is stronger when the anti pi bond is removed (because the donation into its antibonding orbital is no longer present and hence the bond is stronger).

Q4. I think you might have slightly misunderstood the object of this question. You were supposed to calculate energies for A and B as well as conformational isomers A’ and B’ which have the nucleophilic acetyl group twisted so that its orientation with respect to the oxonium group is the opposite. It doesn’t make sense to compare the reactivity of A with that of B to determine a reaction outcome because they are different compounds giving two different products (C and D) and the pathways can never interchange because A cannot convert into B (at least in the scope of this question). If there is any similar confusion in the future modules please ask a demonstrator. In general, the choice of methyl for the R group is the best one to go for and the semi-empirical method is the best to use. You can see from your energy values that A=C and B=D for the PM6 calculations – this is because MOPAC treats the structure as a non-classical carbocation, incorporating the neighbouring group effect into a hybrid structure.

MP. This is a very thorough mini-project and shows how NMR prediction can be very accurate for conformational restricted compounds. You have given all of your calculated and literature data and I can see on analysing it that there is a good correlation, however it could have been presented in a more explicit way. Even stating the differences in ppm in the table would be good, but better would be a graphical display (bar chart is often used) of these differences to see at a glance the general strength of the correlation. Further than this there are countless error analysis methods that could be applied. It would also have been interested to see how the data calculated for isomers of the molecules analysed compared to the lit data for different structures. If you could show that the lit data is closer to the calculated data for the real structure than for other possible isomers then it would be possible to use calculated NMR spectra as a diagnostic tool. Did you find out how exactly the different isomers were distinguished experimentally. There are numerous 2D NMR techniques for doing so and an explanation of this would have been complimentary to the investigation. The IR data is a surprisingly good match – as you say it has associated inaccuracies (in part because it is calculated in the gas phase). Also your optical rotations all match the experimentally observed sign which is perhaps the most important aspect of the technique.

Talk:Mod:HAS1501

2011-10-27T10:31:23Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your calculated values are all good, although there is no need to include so many significant figu..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your calculated values are all good, although there is no need to include so many significant figures: There is some error in the calculation which is after all an approximation and you have probably seen using ChemBio3D that the latter decimal figures won’t be the same if you do the same calculation twice from scratch. The major difference between compounds 3 and 4 is correctly identified as bending strain – for a more complete answer, the structures could be compared to find the origin of this difference, namely the difference bond angles in the double bonds present. Your comment on the hydrogenation reaction is right – under thermodynamic control you would get 4 and in order to get compound 3 the reaction would have to be under kinetic control.

Q2. Your calculated structures are not the lowest energy forms although isomer B is indeed the lowest energy one. Both of your compounds, in fact, have the 6-ring in the twist-boat form. In the lowest energy form the 6-rings should be in the chair conformation. It would have been worth detailing briefly the measures you took to minimise the structures, an analysis of the different possibilities for the 6-ring would have lead you to the right answer. The discussion of the strain contributions is good. Hyperstable alkenes are well defined, but it is not possible to carry out that comparison of the hydrogenated “parent” compounds using molecular mechanics, because the compounds are not isomeric. It would be possible using the absolute scale of a DFT calculation.

Q3. Your MOs look good, as do the calculated IR spectra. The energies obtained for the main compound with the different calculation methods are not reported – in the future modules make sure you report these different energy values if appropriate, as you may be assessed on the numbers. You have identified the possibility of a pi-sigma* interaction, but did not discuss the impact this has on your various IR stretches. The anti pi bond has a lower energy stretch than the syn pi bond and the C-Cl stretch is stronger when the anti pi bond is removed (because the donation into its antibonding orbital is no longer present and hence the bond is stronger).

Q4. I think you might have slightly misunderstood the object of this question. You were supposed to calculate energies for A and B as well as conformational isomers A’ and B’ which have the nucleophilic acetyl group twisted so that its orientation with respect to the oxonium group is the opposite. It doesn’t make sense to compare the reactivity of A with that of B to determine a reaction outcome because they are different compounds giving two different products (C and D) and the pathways can never interchange because A cannot convert into B (at least in the scope of this question). If there is any similar confusion in the future modules please ask a demonstrator. In general, the choice of methyl for the R group is the best one to go for and the semi-empirical method is the best to use. You can see from your energy values that A=C and B=D for the PM6 calculations – this is because MOPAC treats the structure as a non-classical carbocation, incorporating the neighbouring group effect into a hybrid structure.

MP. This is a very thorough mini-project and shows how NMR prediction can be very accurate for conformational restricted compounds. You have given all of your calculated and literature data and I can see on analysing it that there is a good correlation, however it could have been presented in a more explicit way. Even stating the differences in ppm in the table would be good, but better would be a graphical display (bar chart is often used) of these differences to see at a glance the general strength of the correlation. Further than this there are countless error analysis methods that could be applied. It would also have been interested to see how the data calculated for isomers of the molecules analysed compared to the lit data for different structures. If you could show that the lit data is closer to the calculated data for the real structure than for other possible isomers then it would be possible to use calculated NMR spectra as a diagnostic tool. Did you find out how exactly the different isomers were distinguished experimentally. There are numerous 2D NMR techniques for doing so and an explanation of this would have been complimentary to the investigation. The IR data is a surprisingly good match – as you say it has associated inaccuracies (in part because it is calculated in the gas phase). Also your optical rotations all match the experimentally observed sign which is perhaps the most important aspect of the technique.

Talk:Mod:BMWWiki

2011-10-27T10:30:36Z

Mjhughes:

FEEDBACK

Q1: Your energy values are correct and the data is well presented. You should make sure you check the compound numbers from the course wiki because your monohydrogenated compounds 3 and 4 are misassigned. It is correct to say that the lower energy compounds are the ones you would expect to obtain under thermodynamic control. The major difference between the monohydrogenated compounds is correctly identified as the bending strain due to deviation from ideal sp2 angle. Just to clarify this corresponds to the bond angle (angle between 3 consecutive atoms) not the dihedral angle (angle between 3 consecutive bonds).

Q2. Your calculations are good. It makes sense to analyse the 6-ring as an obvious site that could have a different conformations. It would have been good to hear about other ways you attempted to optimise the structure (successful or unsuccessful). With the stabilising orbital interaction – I presume you mean an interaction between a C-H sigma bond and the carbonyl pi*, it does look like a better overlap in the structure of isomer 10. (NB: When you draw a CH sigma orbital you should make sure that one lobe is smaller than the other – it is an sp3 orbital not a p orbital). Of course, this contribution cannot be analysed by MM2 calculations which do not incorporate molecular orbital interactions – Perhaps a semi-empirical method would have shown a much larger difference in energy between the isomers due to this. The definition of a hyperstable alkene is spot on.

Q3. The structures look good and the MOs and IR stretches are correct, but you should have stated the energies of the compounds you obtained with the different calculation methods as it is hard to make an overall assessment on the calculation without this detail. The syn double bond is indeed the most reactive towards electrophilic attack. The description of the pi-sigma* orbital interaction and its implications for the IR spectrum of the different compounds is done well.

Q4. Your energy values are good – mostly spot on, others close to expected. I think that the methyl group is the best choice here given a generic alkyl chain and semi-empirical methods are much better for this type of system. Your energies and jmols show that for the MOPAC calculations A=C and B=D. That is the calculation cannot distinguish between them because the cation is treated as a non-classical carbocation and the neighbouring group effect is incorporated into a hybrid structure. This reaction is an example of Curtin-Hammett kinetics, the selectivity is defined by the relative amounts of the isomeric starting materials and the inherent reactivity of each isomer. In this case both factors favour formation of C and D: A and B are lower in energy than A’ and B’ and have a better trajectory for nucleophilic attack.

MP. The task set out at the beginning of this mini-project is ambitious but is exactly the type of question that should be attempted here. Calculation of NMR data for a set of diastereomers and then comparison of that data to a set of genuine data for one of those compounds is a good way to test whether a predicted NMR spectrum can be used to distinguish different isomers. Here you state that the data seems to match the intended compound, but it would have been nice to see some error analysis and a more quantitative approach to making that assessment. The IR assignment is a good extra thing to analyses, however it is unlikely to allow for distinguishing between different isomers because as you stated the calculations aren’t very accurate – for a start they are calculated in the gas phase, whereas in reality they are usually obtained in the liquid or solid state. Did you consider how you could tell these isomers apart experimentally (without calculations); it should be possible to envisage NOESY NMR experiments which can show the proximity of various proton present in the structures.

Talk:Mod:BMWWiki

2011-10-27T10:29:50Z

Mjhughes:

[[FEEDBACK]]

Q1: Your energy values are correct and the data is well presented. You should make sure you check the compound numbers from the course wiki because your monohydrogenated compounds 3 and 4 are misassigned. It is correct to say that the lower energy compounds are the ones you would expect to obtain under thermodynamic control. The major difference between the monohydrogenated compounds is correctly identified as the bending strain due to deviation from ideal sp2 angle. Just to clarify this corresponds to the bond angle (angle between 3 consecutive atoms) not the dihedral angle (angle between 3 consecutive bonds).

Q2. Your calculations are good. It makes sense to analyse the 6-ring as an obvious site that could have a different conformations. It would have been good to hear about other ways you attempted to optimise the structure (successful or unsuccessful). With the stabilising orbital interaction – I presume you mean an interaction between a C-H sigma bond and the carbonyl pi*, it does look like a better overlap in the structure of isomer 10. (NB: When you draw a CH sigma orbital you should make sure that one lobe is smaller than the other – it is an sp3 orbital not a p orbital). Of course, this contribution cannot be analysed by MM2 calculations which do not incorporate molecular orbital interactions – Perhaps a semi-empirical method would have shown a much larger difference in energy between the isomers due to this. The definition of a hyperstable alkene is spot on.

Q3. The structures look good and the MOs and IR stretches are correct, but you should have stated the energies of the compounds you obtained with the different calculation methods as it is hard to make an overall assessment on the calculation without this detail. The syn double bond is indeed the most reactive towards electrophilic attack. The description of the pi-sigma* orbital interaction and its implications for the IR spectrum of the different compounds is done well.

Q4. Your energy values are good – mostly spot on, others close to expected. I think that the methyl group is the best choice here given a generic alkyl chain and semi-empirical methods are much better for this type of system. Your energies and jmols show that for the MOPAC calculations A=C and B=D. That is the calculation cannot distinguish between them because the cation is treated as a non-classical carbocation and the neighbouring group effect is incorporated into a hybrid structure. This reaction is an example of Curtin-Hammett kinetics, the selectivity is defined by the relative amounts of the isomeric starting materials and the inherent reactivity of each isomer. In this case both factors favour formation of C and D: A and B are lower in energy than A’ and B’ and have a better trajectory for nucleophilic attack.

MP. The task set out at the beginning of this mini-project is ambitious but is exactly the type of question that should be attempted here. Calculation of NMR data for a set of diastereomers and then comparison of that data to a set of genuine data for one of those compounds is a good way to test whether a predicted NMR spectrum can be used to distinguish different isomers. Here you state that the data seems to match the intended compound, but it would have been nice to see some error analysis and a more quantitative approach to making that assessment. The IR assignment is a good extra thing to analyses, however it is unlikely to allow for distinguishing between different isomers because as you stated the calculations aren’t very accurate – for a start they are calculated in the gas phase, whereas in reality they are usually obtained in the liquid or solid state. Did you consider how you could tell these isomers apart experimentally (without calculations); it should be possible to envisage NOESY NMR experiments which can show the proximity of various proton present in the structures.

Talk:Mod:BMWWiki

2011-10-27T10:29:17Z

Mjhughes: Created page with " Normal 0 false false false EN-GB X-NONE X-NONE Q1: Your energy values are correct and the data is well presented. You should make sure you check the c..."

Normal 0 false false false EN-GB X-NONE X-NONE

Q1: Your energy values are correct and the data is well presented. You should make sure you check the compound numbers from the course wiki because your monohydrogenated compounds 3 and 4 are misassigned. It is correct to say that the lower energy compounds are the ones you would expect to obtain under thermodynamic control. The major difference between the monohydrogenated compounds is correctly identified as the bending strain due to deviation from ideal sp2 angle. Just to clarify this corresponds to the bond angle (angle between 3 consecutive atoms) not the dihedral angle (angle between 3 consecutive bonds).

Q2. Your calculations are good. It makes sense to analyse the 6-ring as an obvious site that could have a different conformations. It would have been good to hear about other ways you attempted to optimise the structure (successful or unsuccessful). With the stabilising orbital interaction – I presume you mean an interaction between a C-H sigma bond and the carbonyl pi*, it does look like a better overlap in the structure of isomer 10. (NB: When you draw a CH sigma orbital you should make sure that one lobe is smaller than the other – it is an sp3 orbital not a p orbital). Of course, this contribution cannot be analysed by MM2 calculations which do not incorporate molecular orbital interactions – Perhaps a semi-empirical method would have shown a much larger difference in energy between the isomers due to this. The definition of a hyperstable alkene is spot on.

Q3. The structures look good and the MOs and IR stretches are correct, but you should have stated the energies of the compounds you obtained with the different calculation methods as it is hard to make an overall assessment on the calculation without this detail. The syn double bond is indeed the most reactive towards electrophilic attack. The description of the pi-sigma* orbital interaction and its implications for the IR spectrum of the different compounds is done well.

Q4. Your energy values are good – mostly spot on, others close to expected. I think that the methyl group is the best choice here given a generic alkyl chain and semi-empirical methods are much better for this type of system. Your energies and jmols show that for the MOPAC calculations A=C and B=D. That is the calculation cannot distinguish between them because the cation is treated as a non-classical carbocation and the neighbouring group effect is incorporated into a hybrid structure. This reaction is an example of Curtin-Hammett kinetics, the selectivity is defined by the relative amounts of the isomeric starting materials and the inherent reactivity of each isomer. In this case both factors favour formation of C and D: A and B are lower in energy than A’ and B’ and have a better trajectory for nucleophilic attack.

MP. The task set out at the beginning of this mini-project is ambitious but is exactly the type of question that should be attempted here. Calculation of NMR data for a set of diastereomers and then comparison of that data to a set of genuine data for one of those compounds is a good way to test whether a predicted NMR spectrum can be used to distinguish different isomers. Here you state that the data seems to match the intended compound, but it would have been nice to see some error analysis and a more quantitative approach to making that assessment. The IR assignment is a good extra thing to analyses, however it is unlikely to allow for distinguishing between different isomers because as you stated the calculations aren’t very accurate – for a start they are calculated in the gas phase, whereas in reality they are usually obtained in the liquid or solid state. Did you consider how you could tell these isomers apart experimentally (without calculations); it should be possible to envisage NOESY NMR experiments which can show the proximity of various proton present in the structures.

User talk:Xx108 (module 1)

2011-10-27T10:27:53Z

Mjhughes: /* feedback */

==feedback==

Q1: Your energy values are good – although you should have shown at least images of your optimised structures! Too many jmols can cause the wiki page to crash, but jpgs are no problem. Also, it is not necessary to quote values to so many decimal places – There is error to account for in the calculation which is an estimation to start with. You are right to say that the lowest energy exo-dimer is the thermodynamic product and the endo-dimer is the kinetic product. It should be noted however that it is NOT the kinetic product because it is the highest in energy (all that means is that it isn’t the thermodynamic product). The kinetic product of a reaction is the one which has the lowest energy transition state that leads to it. In this case it so happens that the endo-dimer is higher in energy but has a lower energy transition state due to the secondary orbital interaction you mentioned. Typically, the kinetic product is the same as the thermodynamic product (i.e. lowest energy product comes from the lowest energy transition state). For the mono hydrogenated dimer the analysis of strain contributions is good – the major difference is indeed due to that double bond in the bicycle deviating from the ideal sp2 bond angles more than the one in the 5-ring. Again, higher energy product is not synonymous with kinetic product: In this case it isn’t really possible to comment on which product is formed by hydrogenation without considering the conditions employed. For the reduction of a double bond there are a variety of different conditions, mostly involving a metal catalyst and hydrogen gas – usually these reactions are irreversible and kinetically controlled. Without calculating transition states (not possible with MM2), it is not possible to make computational assessments of selectivity.

Q2. Again, some pictures or jmols would be useful. Are you sure you mean boat and not twist-boat? The boat conformation of a 6-ring is usually much higher in energy than chair or twist-boat. Your optimised energies are spot on and the lowest energy atropisomer is correctly assigned to be the down form. Trying different forms of the 6-ring is a good main focus for this question, but you could have commented a bit more on what else you tried to do in order to get the lowest energy. A comparison of the strain contributions would also have been good. The definition of a hyperstable alkene is good.

Q3. Did you attempt the symmetrisation in GaussView to avoid asymmetrical orbitals? As you correctly pointed out they should be symmetrical, but this calculation can be unreliable using lower level computational methods. You should give your energy values for all calculation methods so that the energies can be assessed. Your description of the pi-sigma* interaction is a little confused; the stretching frequency of the C-Cl bond increases when the double bond is removed, indicating a stronger bond because there is no longer donation into the sigma*. You can also see the impact of this interaction in the C=C stretching frequencies – lower for the anti double bond which donates its bonding electrons and is hence weakened.

Q4. The methyl group is a good choice for this calculation and in general for computational chemistry on molecules with generic alkyl groups. Also I agree that semi-empirical methods are better in this case. You should have found that for MOPAC calculations A=C and B=D and the cation is treated more as a non-classical carbocation. It is correct that all of A’/B’/C’/D’ are higher in energy than A/B/C/D. Your actual energy values are a little high and it is difficult to work out why without some jmols to look at. The reason the reactions are diastereoselective is in part due to the dominance of A and B over A’ and B’ and also due to unfavourable orientation of A’ and B’ for nucleophilic attack on the oxonium carbon.

MP. You are right to that you can monitor the reaction by the disappearance of the C=O stretch in the IR spectrum, but optical rotation would not be the first choice for distinguishing alcohols 6 and 7. In this case 2D NMR techniques could be employed such as NOESY which could be used to show the proximity of the hydrogen next to the alchol group to other hydrogens in the molecule. In your analysis of the NMR data, it is better to compare the differences in ppm rather than % differences. The use of ppm makes this analysis arbitrarily smaller for high field (low ppm) resonances. If you considered the actually frequencies involved in MHz, the percentage differences would be much different. Generally, 2-3 ppm is seen as a large error in an NMR calculation, so your values are not too bad. The key question is whether this error is small enough that you could distinguish between the two possible isomers. You could have calculated the spectrum for the other alcohol and compared those values to the reported data – hopefully it would be significantly different and you could therefore use calculated NMR to tell the two compounds apart. In order to discuss the selectivity of the reaction, it would have been useful to see a scheme illustrating the reaction mechanism; you suggest that trapping of an intermediate with tBuOH is stereodetermining – I presume this means protonation of an initially formed radical anion, but this is not clear.

==jytyty==

jiui

===jkkucp===

uyiuyl

==ytiy==

oiuio

===iliou===

uyool