Assignment of Series 4 NMR Spectra

[mattodd]

Tom - I see your assignment work here:

http://malaria.ourexperiment.org/triazolopyrazine_se/10183

Very nice, well done. Question: has anyone checked through your assignment already? We ought to double-check it, since I can see this being used as a very useful check for a number of compounds we'll make in the future - i.e. this is one of those things we need to be really sure about before we build on it too much. Maybe we can put this one out to the Twittersphere and see what we get? If not, then I'll check it ;).

[tscmacdonald]

I don't think anyone else has checked them over in detail, but I went over them with Ian to remind myself of how to interpret HMBC spectra and nothing seemed to jump out at him. Have a look over it if you can, and I'll run some HMBCs on my other analogues to see if everything looks similar (it seems like the right things have moved and stayed still on my other 13C's, but it's hard to tell. With the number of aromatic signals I have, it's not hard to find signals when looking).

[mattodd]

Well, let's try this. You'll need to write up your assignment in one paragraph in your Honours thesis, pointing out the key signals you used, and your thought process. You'll need one nice diagram - Kat made some lovely ones for her thesis. The same diagram can go in your talk. I suggest you write the text and make the diagram. Then we post those - can go on the ELN - and ask people for input/checking. So your audience is a PhD-level organic chemist, unaware of OSM who is short on time but big in heart who might want to assist. Try to do this now, while you remember your thought process. It means you will have written up approximately 0.7% of your thesis in the process, so everyone wins!

On 30 June 2014 03:25, tscmacdonald notifications@github.com wrote:

I don't think anyone else has checked them over in detail, but I went over them with Ian to remind myself of how to interpret HMBC spectra and nothing seemed to jump out at him. Have a look over it if you can, and I'll run some HMBCs on my other analogues to see if everything looks similar (it seems like the right things have moved and stayed still on my other 13C's, but it's hard to tell. With the number of aromatic signals I have, it's not hard to find signals when looking).

— Reply to this email directly or view it on GitHub https://github.com/OpenSourceMalaria/OSM_To_Do_List/issues/224#issuecomment-47501800 .

MATTHEW TODD | Associate Professor School of Chemistry | Faculty of Science

THE UNIVERSITY OF SYDNEY Rm 519, F11 | The University of Sydney | NSW | 2006 T +61 2 9351 2180 | F +61 2 9351 3329 | M +61 415 274104 E matthew.todd@sydney.edu.au | W http://sydney.edu.au/science/chemistry/research/todd.html | W http://opensourcemalaria.org/

CRICOS 00026A This email plus any attachments to it are confidential. Any unauthorised use is strictly prohibited. If you receive this email in error, please delete it and any attachments.

[egonw]

You could run it via http://NMRShiftDB.org/ but I am not sure if the data in the repository is good enough to distinguish this solution from the alternative possible structures.... (either case, you should consider submitting your structures and peak assignments there anyway!)

[tscmacdonald]

Thank you! I'm not sure how I wasn't aware of that database, but everything seems very close to the values I've found. I'll try to add structures to there from now on. The SMILES is O=C(NC1=CC=CC(Cl)=C1)C2=CN=CC3=NN=C(C4=CC=C(OC(F)F)C=C4)N32 if anyone wants to try for themself. My ELN post has a few paragraphs on rationalisation now (Mat: I couldn't find anything very comparable in Kat's thesis).

[mattodd]

Is the numbering of atoms in your chloroanilide section correct, i.e. 11 and 13? And are you just seeking more certainty on those remaining two C's? In the gap you have there under the word "Rationale" mention that these molecules are new, so the NMR spectra have not, as far as you know, been properly assigned before. Highlight the fact that you think you have assigned everything, but there remains a single uncertainty. Then when edited, let's tweet it out.

On 1 July 2014 01:13, tscmacdonald notifications@github.com wrote:

Thank you! I'm not sure how I wasn't aware of that database, but everything seems very close to the values I've found. I'll try to add structures to there from now on. The SMILES is O=C(NC1=CC=CC(Cl)=C1)C2=CN=CC3=NN=C(C4=CC=C(OC(F)F)C=C4)N32 if anyone wants to try for themself. My ELN post has a few paragraphs on rationalisation now (Mat: I couldn't find anything very comparable in Kat's thesis).

— Reply to this email directly or view it on GitHub https://github.com/OpenSourceMalaria/OSM_To_Do_List/issues/224#issuecomment-47617738 .

MATTHEW TODD | Associate Professor School of Chemistry | Faculty of Science

THE UNIVERSITY OF SYDNEY Rm 519, F11 | The University of Sydney | NSW | 2006 T +61 2 9351 2180 | F +61 2 9351 3329 | M +61 415 274104 E matthew.todd@sydney.edu.au | W http://sydney.edu.au/science/chemistry/research/todd.html | W http://opensourcemalaria.org/

CRICOS 00026A This email plus any attachments to it are confidential. Any unauthorised use is strictly prohibited. If you receive this email in error, please delete it and any attachments.

[tscmacdonald]

Well caught: it seem chemdraw moved all the atom numberings around between my hand-annotated printout, and the figure there. All the numbers were a mixture between the two schemes (now fixed).

[PatrickThomson]

I've downloaded the raw data without priming myself with @tscmacdonald's interpretation (too much); I'll see what I make of it. I was going to suggest through-space experiments but that wouldn't give us a whole lot more data in this structure IMO.

[PatrickThomson]

We have institutional access to ACD/I-Lab, which has a fairly decent NMR predictor built into it. That, plus literature, both suggest that in the 3-chloroacetanilide motif, the "ortho" carbon is 6ppm downfield compared to the "para" carbon -

Prediction:

Literature for N-(3-chlorophenyl)benzamide (Compound 1C, Table 3, 1C to 6C refer to the "aniline" ring as per key on para 2 p 1222) http://onlinelibrary.wiley.com/doi/10.1002/jhet.5570390616/pdf

Everything else about the assignment looks good so far - the enormous 1H-19F coupling threw me off for a bit though!

[edit] Two of the heteroatoms are missing from the center ring of the predictor - I had to redraw it, but the shifts were about the same beforehand).

[mattodd]

Looks good. Patrick your aromatic core lacks 2 nitrogens but I'd wager that makes no difference. @tscmacdonald like it?

On 7 July 2014 09:28, PatrickThomson notifications@github.com wrote:

We have institutional access to ACD/I-Lab, which has a fairly decent NMR predictor built into it. That, plus literature, both suggest that in the 3-chloroacetanilide motif, the "ortho" carbon is 6ppm downfield compared to the "para" carbon -

Prediction: [image: image] https://cloud.githubusercontent.com/assets/4561884/3495593/d20f9b28-05d8-11e4-926b-fd86f5cbc175.png

Literature for N-(3-chlorophenyl)benzamide (Compound 1C, Table 3, 1C to 6C refer to the "aniline" ring as per key on para 2 p 1222) http://onlinelibrary.wiley.com/doi/10.1002/jhet.5570390616/pdf

Everything else about the assignment looks good so far - the enormous 1H-19F coupling threw me off for a bit though!

— Reply to this email directly or view it on GitHub https://github.com/OpenSourceMalaria/OSM_To_Do_List/issues/224#issuecomment-48177645 .

MATTHEW TODD | Associate Professor School of Chemistry | Faculty of Science

THE UNIVERSITY OF SYDNEY Rm 519, F11 | The University of Sydney | NSW | 2006 T +61 2 9351 2180 | F +61 2 9351 3329 | M +61 415 274104 E matthew.todd@sydney.edu.au | W http://sydney.edu.au/science/chemistry/research/todd.html | W http://opensourcemalaria.org/

CRICOS 00026A This email plus any attachments to it are confidential. Any unauthorised use is strictly prohibited. If you receive this email in error, please delete it and any attachments.

[PatrickThomson]

I've just re-checked the structure and it's not smart enough to go that far away - the values are exactly the same. ACD/I-Lab is infuriating in that it gives me a comprehensive PDF report of the NMR prediction, but refuses to automatically number the atoms for me - I need to mouse-hover each atom individually to get the results. (or, I'll bet, use ACD's own structure editor...)

[PatrickThomson]

It (very non-obviously) uses the atom numbering from the ChemDraw I pasted in. Here are the reports it generated for me, in ascending order of usefulness:

IUPAC name: https://drive.google.com/file/d/0B6rAjqmPhHy5dVlieVZJYThnb1E/edit?usp=sharing PK properties: https://drive.google.com/file/d/0B6rAjqmPhHy5dEg1NHlwdHhGNFU/edit?usp=sharing 13C NMR: https://drive.google.com/file/d/0B6rAjqmPhHy5ZXNsVGNnbjJiWE0/edit?usp=sharing

It gets very sketchy predicting the TP ring NMR; the non-quats we have HSQC for are all over the place, but it's pretty good for aromatics, aliphatics, etc.

[mattodd]

Great, that's useful, Patrick, excellent input. Tom, make sure all this analysis is properly cross-linked to the lab notebook and that we have a self-consistent picture. As I said I think that once you fully assign one of these guys, others working on series 4 will use that as a reference point.

On 7 July 2014 14:16, PatrickThomson notifications@github.com wrote:

It (very non-obviously) uses the atom numbering from the ChemDraw I pasted in. Here are the reports it generated for me, in ascending order of usefulness:

IUPAC name:

https://drive.google.com/file/d/0B6rAjqmPhHy5dVlieVZJYThnb1E/edit?usp=sharing PK properties:

https://drive.google.com/file/d/0B6rAjqmPhHy5dEg1NHlwdHhGNFU/edit?usp=sharing 13C NMR:

https://drive.google.com/file/d/0B6rAjqmPhHy5ZXNsVGNnbjJiWE0/edit?usp=sharing

It gets very sketchy predicting the TP ring NMR; the non-quats we have HSQC for are all over the place, but it's pretty good for aromatics, aliphatics, etc.

— Reply to this email directly or view it on GitHub https://github.com/OpenSourceMalaria/OSM_To_Do_List/issues/224#issuecomment-48218056 .

MATTHEW TODD | Associate Professor School of Chemistry | Faculty of Science

THE UNIVERSITY OF SYDNEY Rm 519, F11 | The University of Sydney | NSW | 2006 T +61 2 9351 2180 | F +61 2 9351 3329 | M +61 415 274104 E matthew.todd@sydney.edu.au | W http://sydney.edu.au/science/chemistry/research/todd.html | W http://opensourcemalaria.org/

CRICOS 00026A This email plus any attachments to it are confidential. Any unauthorised use is strictly prohibited. If you receive this email in error, please delete it and any attachments.

[tscmacdonald]

Commenting to confirm that I've seen this + will act on it ASAP, but as I'm currently with the family in Brisbane it won't be until Friday or so. Thanks for the analysis, Pat: everything seems in agreement with what I've said, which is always reassuring. I've already done a quick assignment of TM 19 (similar to this, but with a 3-chloro-2-fluoro anilide, instead of the 3-chloro here: lots of C-F splitting, making it a bit hard to work out where I have two signals, and where doublets)) using this template, which should be tidied up and posted soonish.

[tscmacdonald]

An update to this! I now have fully assigned spectra for each step of our current reaction scheme, starting from the representative 3-chloro anilide. I've also included the bare chlorotriazolopyrazine I've synthesised recently (in low yield). Interesting things to note: the chemical shifts of the aryl amide barely move at all over the course of the synthesis, and there's only a small difference between the hydrazone intermediate and cyclised triazolopyrazine for the "north-east" aryl peaks. @drc007: I think this is evidence against the aryl-aryl interactions you were asking about earlier, but I'd like to hear your thoughts on the matter.

[alintheopen]

Lovely work Tom! Lets get this written up and in a nice scheme for your thesis - how satisfying. :)

On Mon, Aug 11, 2014 at 2:26 PM, tscmacdonald notifications@github.com wrote:

An update to this! I now have fully assigned spectra for each step of our current reaction scheme, starting from the representative 3-chloro anilide. [image: all assignments] https://cloud.githubusercontent.com/assets/6529539/3872078/f4d396e0-210e-11e4-8552-f75ad83e4821.png I've also included the bare chlorotriazolopyrazine I've synthesised recently (in low yield). Interesting things to note: the chemical shifts of the aryl amide barely move at all over the course of the synthesis, and there's only a small difference between the hydrazone intermediate and cyclised triazolopyrazine for the "north-east" aryl peaks. @drc007 https://github.com/drc007: I think this is evidence against the aryl-aryl interactions you were asking about earlier, but I'd like to hear your thoughts on the matter.

— Reply to this email directly or view it on GitHub https://github.com/OpenSourceMalaria/OSM_To_Do_List/issues/224#issuecomment-51741006 .

[drc007]

@tscmacdonald, @alintheopen, @mattodd Very nice work, actually I think this is evidence for an aryl-aryl interaction. If you look at the shifts on the 3-chorophenyl ring. If you look at the shift of the proton at the 4-position it is virtually unchanged between the acyclic (7.2 - 7.22) and the cyclised structure (7.15). In contrast if you look at the ortho protons, these are around 7.8 and 8.0 in the acyclic structures, but are 7.17 and 7.4 in the cyclised structures. A reasonable explanation might be they are shifted upfield due to the shielding effect of the other benzene ring. Is it feasible to get a X-ray structure to confirm?

[tscmacdonald]

...now you point that out, it seems blindingly obvious (except, apparently, to me). X-ray structures should certainly be possible; I haven't needed any yet this year, but you're right that it might be interesting here. I assume we'd be looking at the way the two rings lie next to each other, then avoiding substituents that would disturb that?

[drc007]

@tscmacdonald Precisely.

[mattodd]

Sounds very good, Tom. You have to write this up in a concise form, giving reasoning for any choices you made. Since this is needed for the thesis, why not do this now, so we can check it and work out if any more experiments are needed. If you already have this reasoning on various pages, just link out to them. Crystal structures - set up a few attempts in parallel - ask around if you're not sure how. Should be a breeze to get a nice crystal from these guys, and now there is a good reason to do so.

Maybe there's a nice visual way of representing the changes Chris was talking about. An X-axis with carbon/hydrogen position number, then a y-axis that shows the change in ppm value, colour-coded. That would flag up quickly what's changing and what's not.