Lecture 21. Using HMBC to Help Solve Structures
This is a modal window.
Das Video konnte nicht geladen werden, da entweder ein Server- oder Netzwerkfehler auftrat oder das Format nicht unterstützt wird.
Formale Metadaten
| Titel |
| |
| Untertitel |
| |
| Alternativer Titel |
| |
| Serientitel | ||
| Teil | 21 | |
| Anzahl der Teile | 29 | |
| Autor | ||
| Lizenz | CC-Namensnennung - Weitergabe unter gleichen Bedingungen 3.0 USA: Sie dürfen das Werk bzw. den Inhalt zu jedem legalen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen und das Werk bzw. diesen Inhalt auch in veränderter Form nur unter den Bedingungen dieser Lizenz weitergeben. | |
| Identifikatoren | 10.5446/19261 (DOI) | |
| Herausgeber | ||
| Erscheinungsjahr | ||
| Sprache |
Inhaltliche Metadaten
| Fachgebiet | ||
| Genre | ||
| Abstract |
|
00:00
Chemische StrukturLactoseAlkoholische LösungWeich-PVCDigoxigeninKernreaktionsanalyseAschePilleKohlenstofffaserIonenbindungAzokupplungHydrocarboxylierungChemische StrukturWasserstoffMesomerieDeprotonierungStickstoffatomOrdnungszahlSystemische Therapie <Pharmakologie>MutationszüchtungWerkzeugstahlMethylgruppeOxygenierungMolekülAlkeneChemische FormelEsterCobaltoxideTellerseparatorHeteroatomare VerbindungenDoppelbindungDiederwinkelElektrostahlFlammeBiologisches MaterialAllmendeSäureSetzen <Verfahrenstechnik>StromschnelleEnantiomereRacemisierungSingle electron transferStereochemieChloroformWasserSingulettzustandTrennverfahrenChiralität <Chemie>Alkohole <tertiär->Heck-ReaktionHärteAmineEmissionsspektrumValenz <Chemie>Bukett <Wein>RöntgenspektrometerFülle <Speise>BindegewebeFunktionelle GruppeZündholzGletscherzungeWasserscheideRNS-SyntheseSenseAlkoholische LösungHalogeneZunderbeständigkeitDeformationsverhaltenBindungstheorie <Chemie>CyclopropanEthylgruppeChemische VerschiebungAmine <primär->HalbedelsteinKetoneTopizitätCarboxylierungHydroxylgruppeMultiproteinkomplexAlkylammoniumverbindungenMeeresspiegelEthanPhenylgruppeAlphaspektroskopieQuellgebietMethoxygruppeKomplikationErtrinkenChemischer ProzessArtefakt <Histologie>LagerungDiamantähnlicher KohlenstoffScreeningPyridinWirtsspezifitätScherfestigkeitBegasungElektronische ZigaretteAcetylenIsotopenmarkierungZähigkeitElementenhäufigkeitAtomBrillenglasSiliciuminfiltriertes SiliciumcarbidMetallGermaneAlterungThermoformenAluminiumFärbenMethanolAromatizitätNitrobenzolKrankengeschichteAdamantanMündungEnergiearmes LebensmittelTumorSchmiedeeisenFlussMassenspektrometerQuarz <alpha->StoffpatentEinschnürungBiogasanlageElektrolytische DissoziationChemische ForschungGradingSekundärstrukturPeriodateShuttle-VektorHydrophobe WechselwirkungFeuerPropionaldehydKonkrement <Innere Medizin>ZuchtzielReaktionsgleichungWursthülleWildbachPotenz <Homöopathie>Vorlesung/Konferenz
Transkript: Englisch(automatisch erzeugt)
00:07
Yeah, so as you may have noticed this is from your homework and I want to do this as a demonstration regardless of whether you've already solved this problem. In fact, I hope that you've solved this problem
00:21
because what I'd like you to do is to be able to think really deeply about it and show you my perspective on structure solving and in particular we're going to be talking about using HMBC as a very focused tool to help solve structures and specifically for the problem
00:41
of putting the pieces together. You've seen on the templates that I've given you, I've had this tremendous emphasis or at least I've tried to have this tremendous emphasis on thinking your way through the problem, reading the spectrum, getting the formula, writing out fragments, jotting down what you know, jotting down what you don't
01:01
know if possible and then being able to ask focused questions. HMBC is an incredibly data rich technique. It also has ambiguities to it because we get two bond and three bond couplings. So you've got a huge amount of information, a huge amount of data and what I'd like us to do today is
01:21
to learn how to use it as a focused tool. So I want to zip through the problem from Silverstein at the beginning and then focus on the HMBC angle. So this is problem 8.43 from Silverstein and I want
01:42
to jot down my, you know, I've worked this problem just like you folks and I want to jot down my impressions on working the problem. So I look at this problem and I generally start with, if I have, I mean what's the most useful thing you can get, molecular formula and functional groups.
02:01
That's sort of the most general stuff and we'll be able to get molecular formula in just a moment because we'll actually see all the hydrogens and all the carbons so we basically will get them, get the oxygens by difference. So before we start with the NMR spectra though, looking at the IR spectrum I see what looks like an OH group.
02:23
I see what looks like a carbonyl. At 1728 it's a little high for a typical ketone, right, typical ketones a little bit lower, it might be an ester. I've been emphasizing it's very hard in any sort of level of complexity to get CO single bond stretches associated
02:42
with esters. It's probably hiding right here at 1188 so you sort of get a hint at it but we'll see in a moment there's some other hints that are pointing toward ester. We probably have an alkene, that's probably an alkene CH, it could be an aromatic, we'll see in a moment it's not,
03:01
that's probably an alkene CC bond. All right, mass spec, it's 200 molecular weight in the EI mass spectrum. What I'm going to do before we actually get the formula, so maybe at this point I'll just jot down sort of a question mark on the IR spectrum ester,
03:21
sort of my thinking on this. All right, at this point I want to go and look at the proton NMR, look at the carbon NMR and I've really, really, really been emphasizing this notion of keeping the resonances separate from the atoms. In other words, we know resonances,
03:42
it's something you can see. We're going to be assigning those resonances to the structure as we build the structure and so we have a common language. I go ahead and I simply letter the peaks, A, B, C, it's hard staring into the light, D, E, F, G, H, I,
04:12
and we number the carbon resonances and I'll go 1, 2, 3, that's our chloroform, 4, 5, 6, 7, 8, 9, 10, 11.
04:32
I like to do a good job of working the integrals. My philosophy, every particular integral has experimental error
04:41
associated with it. The best way that you can do it if you know the number of hydrogens in the molecule or you know multiple hydrogens is add up a bunch and divide by the number of hydrogens. If you don't or you're in a hurry, often you can get it by inspection with a ruler. These integrals are a little hard to read. They're starting them right in the baseline.
05:01
They're not offset. You can use a ruler and draw straight lines. You can slap a grid on it. I'm just using my ever useful grid and if I'm looking, I start to see a ratio that's just a hair under 6, a hair under 6, a hair under 3, hair under 3, hair under 3.
05:23
This one's just a little bit, actually this is just a hair under 6, hair under 9, so basically we're talking about like 2.8 units. These happen to be tenths of an inch per hydrogen. This guy is interesting.
05:41
He's coming up a little bit. We'll get this one right at 1.6 PPM so he's coming up over .6 and I'll tell you about that in a second and then we see one that's just a hair under .9 and another that's just a hair under .9. So you pretty much by inspection can go ahead and get and again it's really hard staring in here, 2H for A,
06:05
2H for B, 1H for C, 1H for D, 1H for E, 1H for F, 2H for F, thank you, 3H for G. Now H is interesting,
06:24
1.6, this is in chloroform solution and chloroform water typically shows up at about 1.6 parts per million and so you can kind of see the water peak right over here. So you have this multiplet that's reasonably symmetrical
06:41
and then you have a little bit of water off on the side here. I would go in if I were at my own spectrometer I'd go in and zoom in and get a better look and see but I think you can see the integral just gives a little extra kick up for the water. So H is 2H, I is 3H, J is 3H.
07:08
So in other words our molecular formula has a total of H20 in it. Now it's certainly possible if I had an amine or something that the amine NH might not show up or an alcohol NH.
07:25
So I'm not immediately locking in and this is really important. You have to be keeping your wits about you. Secondary amine NHs, aliphatic secondary amine NHs are terribly hard to see. Alcohols can be broad, alcohols can be with the water peak,
07:40
carboxylic acids can be broad as well and can be disappeared particularly if you're in chloroform versus DMSL due to exchange. If we look at the carbon NMR here, peak number 1 is at about 170, just a hair down of 170, about 173 ppm.
08:02
Where you would expect for an astronaut, where you'd expect for a ketone. We've got a couple of peaks over here. If I want to go ahead at this point I'll say 1 is a quat, 2 is a quat looking at the depth, 3 is a CH2, 4 is a quat, 5 is a CH2, 6 is a CH2, 7 and 8 are all CH2s and then 9,
08:28
10 and 11 are all CH3s. So basically if we do this we see we have C11H20. If we have symmetry in the molecule which you've gotten
08:41
in a couple of homework problems this week, of course you might end up with a count that's lower if your carbons aren't all different or you could have overlapping peaks but symmetry is a typical reason. If you have a plain phenyl group you're going to see 4 carbon peaks but you're going to have 6 carbons. If you have a ring like a cyclopropane ring
09:03
and there's no stereocenter in the molecule you may if you have one point of attachment have 2 methylenes that are the same but if I look here I'm suspecting in alcohol, I'm suspecting in ester, I have 200. If I take away C12, C11 and H20 from that I get O3.
09:25
So I'm reasonably happy at this point. I think I have a molecular formula. Now the next thing I do with a molecular formula is I'll start to, and it's nice to have a scorecard here so I might for example
09:41
over here jot down some thoughts, ester, alcohol, alkene as some of the groups that we're seeing in the molecule. If I want I can calculate degrees of unsaturation. Another very useful thing, remember I don't do it
10:03
by formula I just say C11 would be H24. I'm at H20 so we have 2 degrees of unsaturation. So I can help keep my wits about me. That also tells me if I have a carbonyl and I have an alkene I know I have no rings in the molecule.
10:21
All right now the next place, any questions or thoughts at this point? Does it? Let's see. And this is actually a good point in checking yourself. I have an incredible ability to screw up arithmetic
10:41
on my feet which means go ahead and double check your, all right. C could indeed be the OH.
11:00
So we've got some interesting points here you can already see. So if you're reasonably experienced I'm going to bet anything that we have a stereocenter in the molecule. Now even though I haven't measured the coupling constants, I could measure the coupling constants.
11:20
They very conveniently give me a scale here and you'd get a peak print out. But very conveniently a typical aliphatic CH coupling constant is about 7 hertz. A methyl group next to a methylene is really very typical for this. A methyl group next to just about anything is going to give you about 7 hertz.
11:41
So immediately if I'm eyeballing this even if I don't know I'm at 600 megahertz and I see this triplet here that's obviously a CH3 next to a CH3 next to a CH2. In fact I can write this as one of my fragments here because I think we're pretty obvious at that point
12:01
for this and this here. But if I'm looking at that I see, okay, these guys are leaning into each other. We have an AB pattern. It's obviously a big coupling constant. This separation is at least twice the separation in the triplet. So it's like a 14 hertz or 15 hertz coupling constant.
12:21
That is very typical for a geminal coupling. So in the back of my mind I'm thinking stereocenter, something with a methylene that's diastereotopic. If I have a stereocenter every methylene in the molecule is going to be diastereotopic but usually the ones that are further from the stereocenter behave as if things are coincident.
12:42
So we look here. We have what looks like an ethyl group. It wouldn't have surprised me to see a more complicated coupling pattern for this ethyl group but it doesn't necessarily surprise me to see a less coupling constant. I'm pretty sure this is a less complicated coupling pattern just to see a plain old quartet.
13:01
I'm pretty certain I have an ethyl ester at this point. Pretty confident I have an ethyl ester because that sure looks like an OCH2CH3 off of an ethyl ester but it could be something else that's shifting it down field. All right. At this point it's time to get analytical and where I like to go next is actually the HMQC rather than the COSY.
13:25
As I was saying in discussion the other day, the problem is you're drowning in data in a COSY quite often and a lot of the data isn't particularly useful. You'll have two diastereotopic protons
13:42
and they'll both couple to another two diastereotopic protons or even to one proton and you're getting more information than you need because you'll get that the diastereotopic protons are coupling with each other, okay, but big whoop. You'll get that each one is coupling to one of the protons.
14:02
All right, that gives you some new information but that's redundant and if you have two diastereotopic protons coupled to two diastereotopic protons now you've got lots and lots of data points that basically just tells you you have a CH2 next to a CH2. So in order to help make sense of the data I go next
14:20
to the HMQC and I'm very, very rigorous about trying to transcribe stuff and I'm not super neat but I tend to really try to be analytical so I'm going to copy all of my numbers here to the carbon axis and I'll copy all
14:47
of my letters to the proton axis and at this point,
15:03
again, a grid is extremely useful if you're working on your own spectrometer you might want to do an expansion, you might want to expand this region just so you can get a close look and see what's lining up but if you're having trouble following with your eye
15:20
and things aren't completely obvious, slapping a grid on the spectrum is a very useful way, for example, to see that this peak at 9 here is actually crossing with the singlet so we have two methyls that are right next to each other.
15:41
One of those methyls is a singlet, one of them is a triplet and so you can very nicely see that 11 is crossing with I and 9 is crossing with J and then we can go and 10 is crossing with H I believe
16:03
and again I'm going to cheat a little bit because 10 is crossing with G. I'm going to cheat a little bit because it's hard staring in here and it looks like 7 is crossing with H and then F it looks like is crossing with 8 and let's see, 6 is crossing
16:29
with D and E so 6D and 6E and it looks like 5 is crossing with B, C has no partner and it looks
16:43
like 3 is crossing with A and this information here really becomes my Rosetta Stone. It really becomes the key piece of information that I'm going to be using now in building the structure
17:02
because now that's going to give us all of our easy connectivities and you need to keep your wits about you about chemical shift as well. So obviously if we're talking a carbonyl we're probably talking 1 is a carbonyl. If we're talking about an alkene we're probably talking
17:23
about 2 and 3 being the alkene and it looks like 3 is a quad carbon of an alkene and 2 is a CH2 of an alkene so it looks like we have a gem dimethyl group here. 4 and 5 are interesting.
17:41
Remember the region your carbon protons track pretty well with your proton scale and if you're next to not so well on halogen or nitrogen those tend to be sort of if you're next to halogen or nitrogen those tend to be here but if you're next to an oxygen you usually tend
18:01
to be somewhere around here in the 50 to 90 range depending on whether it's an OCH3 at one extreme or say a carbon that's a quad with an alcohol at another extreme. If you're down further than that it could be 2 oxygens on a carbon or it could be an alkene or it could be a nitrile. Okay, so it looks like we probably have a couple of carbons
18:23
with an oxygen attached to them. Number 5 is a CH2 and number 3 as I said was a CH2, 4 is a quad, 2 is a quad, 1 is a quad.
18:41
Quads and other things that are isolated are going to be problems later on. So if you look at this spectrum we have 7H and 9J that don't show that are close to singlets and we'll see 7H is going to come in.
19:00
We have C that has nothing on it and so we're going to have to deal with him. Okay, now at this point I'm prepared to go to my COSY. This particular COSY the way they plotted it in the book to save space or allow you to give bigger spectra they only plotted the axis on one edge and that's okay
19:23
because you can just bounce up and over and then up. Otherwise I'd just be bouncing up and up to that axis. However you want to do it is fine. All right, so now as I said I'm ready to attack my COSY and the first thing I'm going
19:40
to do is again very slavishly transcribe. So I have 3A and 5B and C and you really don't want to make a mistake at this point, 6D, 6E, 8F, 10G.
20:02
The other nice thing in addition to having it so that we all can discuss things together, the other very nice thing about taking this sort of meticulous systematic approach to this in 9J here, the other thing about taking this sort of meticulous systematic approach is it means that you can set
20:22
down the problem, come back to it later and get your bearings a lot more quickly because it's not just oh we've got a lot of stuff. It's actually oh I've got this resonance and I still don't know where it goes. All right at this point we want to sort of build
20:41
up our fragments and I'm going to look at the COSY. You can use a grid. You can draw straight lines on it, whatever you like. If you want a grid you can do it this way. If you're fancy and you like to work off the online problems in Acrobat the command U if you're using a full version,
21:01
I'm not sure about the reader, command U actually just slaps a grid right on the screen which is useful and you can even, there are parameters in Acrobat to set how fine a grid but it's helpful with particularly with the book problems where they tend to be less generous with the expansion. So at this point I tend to go through and just list all
21:23
of my cross peaks so if I'm working up and I only need to work on one side of the diagonal, it should be pretty symmetrical but if you're uncertain whether anything is noise check if it's on both sides of the diagonal. If you're working on the spectrometer you can go down a level or up a level with the times 2 or divided
21:42
by 2 or the slider and you're going to get down to a level where you have two types of things. At the very bottom you'll just have a basement of noise particularly in data poor experiments but you'll also have artifact baseline role. Your phasing is very important in the 2D experiments in terms
22:01
of not getting a lot of crap off the baseline so you really want to take time when you're phasing it to do a good job. Okay, so we've got this peak 3A to 10G and it looks like we have something of 3A and 8F here.
22:22
I'll put a little question mark there. I'm not completely sure what's going on. Of course I'm sure what's going on but at this point in analyzing the thing I'm not sure. Okay, this is kind of nice because you look here and you say all right, yeah, we've got 6E, 6F but I already know they're on the same carbon. It's no big deal.
22:41
They're diastereotopic protons that are coupled to each other. Okay, I don't need to stress about that. All right, 8F cross 7H and I think that about does it. So at this point I want to put together fragments and this is what I'm using that, oh, here we go.
23:07
Yeah, yeah, yeah, 5B, right. 5B to 11I, absolutely, 5B to 11I and you notice that you're using multiple sorts of tools in thinking
23:25
about things because on the one hand you're recognizing when we just had the 1D data in front of us, you were recognizing the match of patterns and coupling constants. You were recognizing a 3 hydrogen triplet upfield
23:40
and a 2 hydrogen quartet downfield, mid-field probably went together as an ethyl ester group because CH3 split into a triplet has to be next to a CH2. CH2 split into a quartet, remember it didn't have to be a quartet but CH2 split
24:01
into a quartet probably is next to a CH3. It could have been next to a CH2 and a CH where it had the same J value so I don't know for sure but now with this and everything else I think I can be, can be pretty sure so let me sort of jot down what I think I have at this point for my fragment.
24:21
So I think I have OC5H2B, C11H3I as one fragment and I like to use that way of doing it where I put the number on the carbon, the letter on the hydrogen so this just means carbon 5 is part of a CH2 group
24:43
with 2 protons B and C11 is part of a CH3 group with 3 protons I. All right what else do I have in my score card here? Well that C8H2F cross peak with C7H2H gives me another fragment.
25:05
The other thing that I like to do in keeping score at this point is I like to show my valences. It helps my thinking to say okay that means I have a valence here. This helps me think about where my unfilled valences are.
25:24
All right what else do I have in my score card? I have an alkene and it's C3HAHA. If I were thinking about it I might say maybe these are 2
25:44
distinct hydrogens here. If I later went on and I said oh wait I thought they were 1 type of hydrogen but they're 2 and this is important to my analysis all I'd need to do at this point is say all right we're going to call one of them A and one of them A prime
26:01
and the nice thing is you're not renumbering or re-lettering your whole spectrum if you suddenly say oh wait this resonance is really 2 resonances. If I decided that H was really 2 resonances and of course H and F really are 2 resonances because you have a stereocenter and they're methylenes and they're diastereotopic methylenes.
26:21
It's not like the COSY we saw in discussion section the other time where I was going on where we were talking about highly differentiated, highly anisotropic CH2 groups but it's the same sort of deal so if it were important I could say H and H prime, F and F prime if I really started to see distinct cross peaks but honestly I don't
26:41
so this is absolutely fine for the analysis. Okay so I have C3 and then C3 seems to be bound to C2 which is a quat and we have this interesting situation here where we see this nice cross peak off of 10G right?
27:05
10G is a methyl group, 10G is kind of at this position of about 1.7. Now remember I said I like to keep some numbers in my head and my sort of quick and dirty is a benzylic methyl is 2, a methyl that's alpha to a carbonyl is 2 ppm
27:23
and a methyl that's allylic is about 2 ppm and then I said caveat if you really want to keep an extra number in your head keep 1.7 for that allylic methyl. So this is very typical of an allylic methyl. It's exactly what you'd expect for its position and it's exactly the sort of thing that you would expect
27:43
for allylic coupling. So C10H3G comes here and then this kind of makes sense of the cross peak so right we're seeing this cross peak here, we're seeing this little cross peak with 8 so if you want to we'll get it from the HMBC
28:02
if you don't want to do it but if you want to get it at this point it certainly makes a heck of a lot of sense that 3A is crossing with 8 so I can put this in. If you're uncertain about something, if you're uncertain about a connection put it in with a dashed line.
28:22
It's a great tool to remind yourself on your score card on working out the problem if you're worried or concerned about something. So at this point we need to figure out the rest of our stuff so let me continue with our score card. We have a carbonyl and that's our C1.
28:41
I hope you can see this from, yeah, you're just about, just about on the edge of the screen here. All right so we have a carbonyl, we have some other problems here. We have our C6HDHE and I'll keep filling in my valences
29:05
to be good about this just to remind us that we have some issues of valences here. Then I have some carbon and there may be redundancy here. I have some carbon that's connected to C9H3J
29:28
and that carbon has to be a quad. It has to be because we're isolating that methyl over here. I have some other carbon. Remember I have this carbon 4
29:42
and that carbon 4 is pretty far downfield in the C13 NMR spectrum. It's just at about 70 parts per million and that's got to be a carbon next to an oxygen. All right so at this point that's kind of my thinking.
30:04
Now certainly by brute force you could now assemble these pieces. You would basically start to try all different possibilities and eventually you'd probably come up with the right structure. I like this molecule because it's a simple enough molecule
30:23
that you could but what I'd like to show you now is how HMBC really lends this, makes this into a much more systematic let us say process. All right so at this point I'm going to pull
30:43
up the HMBC spectrum and we get a lot of cross peaks. Your textbook as I've railed about has doctored its HMBC spectra so I've literally und doctored the HMBC spectra. What the textbook did which I don't like is it's taken
31:03
out the HMQC type of cross peaks in the spectra. There's no reason to do that. You should be able to identify these. You're going to see them in real data. The HMBC, we've talked about this notion of delays. We've talked about the idea that you're doing what's basically
31:22
a series of pulses and delays where your delays are optimized to different things like coupling constants. So when we talked about our depth spectra last time I said the problem with a depth spectrum is you're choosing, you have to choose a J value for the depth spectrum.
31:41
You're going to choose a typical J value and that's going to give an optimum performance at that J value and guess what? If your J's aren't 145 hertz, if they're a little off, 125 hertz, 160 hertz, you'll be fine but if they're a lot different you'll start to see other things. Now HMBC is optimized to small couplings.
32:03
Small couplings typically mean on the order of 0 to 20 hertz and the spectrum downstairs is sort of centered at 10 which pretty much catches it all but that doesn't mean your 1 bond couplings that are 125 hertz may not sometimes come in and so you can see your 1 bond couplings.
32:23
The pulse sequence, remember in the depth how we turned on the proton decoupler in the final acquisition? In a typical HMBC you're not turning on the proton decoupler in the final part of the acquisition which means you're seeing your CH couplings. Now most of those are 2 bond couplings so most of those are
32:43
on the order of 2 and 3 bond so most of them are on the order of 10 hertz which means, okay, when you're seeing something where your peak's a little bit wide the peak's a little bit wide because you're seeing that coupling but when you're picking up the 1 bond coupling which is not what the experiment's supposed to pick
33:02
up it is extremely obvious. You get these sort of vampire bites around the peak and so I just put a cross through each of them to help remind me of what's going on, right? So this is peak number 11 and this is 11 over here is 11I.
33:24
So those 2 vampire bites there are the 1 bond coupling. It's trivial. That's the information that's already in your HMQC spectrum but it's confusing in this data rich experiment. So let me go through and now systematically number 10, 9, 8, 7, 6, 5, 4, 3, 2,
33:46
1 I'll systematically number my carbon axis. I'll once again slavishly write my labels on the proton axis 3A, 5B, C, 6D, 6E, 8F, 10G, 7H,
34:14
I already did 11I and 9J and you can pretty much see all
34:34
of your sort of vampire bites here. There happen to be 6 pairs of them here. So we're getting 1 around 5.
34:41
That's your HMQC pattern. We're getting 1 around 7. That's your HMQC pattern. We're getting 1 around 6 and we're getting let's see 1 around 10 and 1 around 9.
35:04
All right, even without these types of information you are still extremely data rich and the question becomes where to begin. Now if you're a computer you just go ahead and suck in all the data and fit the pieces together that way but as a person you've got the tremendous strength
35:21
that we've already seen of pattern recognition. We saw it. We could read that ethoxy group in the ester. We could read the diastereotopic methylene and you were seeing that from this whole host of little things with the diastereotopic methylene from the big J value, from the peaks tenting into each other. It's the same way most of you were able
35:42
to tackle those coupling problems on the midterm exam and just read and say okay this is this nitrobenzene, this is that nitrobenzene, this is this methyl pyridine because you were starting to recognize the magnitude of the coupling, the tenting of the peaks and so forth. So you've got some tremendous advantages
36:02
and yet we've also got this ability to or this issue of how do we avoid confusion here. So usually what I like to do are to start with my problems and in general my problems are going to be the isolated peaks. In general it is the peaks that are isolated that are going
36:24
to be both the tough ones to put together like 3, 9 HJ because we don't know where it fits. It's not coupling into anything but they're also going to be the sources of tremendous strength because those isolated ones are then going to link
36:41
to the other isolated parts. So let's start with the carbonyl. The carbonyl is 5B to 1 and 6D and 6E to 1
37:04
and that's very, very useful. So remember HMBC can pick up 2 and 3 bond coupling and the problem is you don't necessarily know which is which. That's the really tough thing with this technique.
37:20
Later on we'll talk about inadequate which is a very powerful technique. It's like a carbon-carbon COSY but not a very useful technique because of the low natural abundance of C13. It's very atypical to have 2 C13s next to each other in a molecule. But that's incredibly powerful because it means you can assemble your whole carbon
37:42
skeleton like a COSY process. But the problem with HMBC is you've got both 2 and 3 so you're always second guessing yourself. So whenever you can avoid second guessing it's a tremendous breath of fresh air. So here the 5B to carbon 1 is very nice
38:04
because that provides a linkage there. We kind of sort of knew this was an ethyl ester but this tells us because the 3 bond coupling is going through there so it can't go any further. There can't be an intervening atom. The 6 is less clear because it could be directly connected
38:24
but at this point I can't know. There could be an intervening atom that's still, we know that the 6 is isolated. We know that D and E are isolated. They're not J coupled to anything else but I don't know about this bond here at this point.
38:40
So at this point I might write a big question mark here. All right, I'm not going to go ad nauseam through everything because a lot of the information is redundant. So for example, here we're seeing that 2 is crossing
39:00
with 7 so this is 8F cross 2, 10G cross 2 and 7H to 2 and in a way you could say all of that is redundant
39:21
because I've already placed groups on here and so your 2, you notice and this is a perfect example, if I weren't already sure of this or pretty sure by allylic coupling I wouldn't necessarily know since 2 is crossing with both 7H
39:42
and 8F I wouldn't necessarily know which end this is going because it's 2 and 3 bond coupling so it could couple either way. So the only thing that's really told me this and I think by this point I can go ahead and turn this into a solid line, the only thing that really told me
40:06
that was the COSY but fortunately we'll get enough other things because we're going to see in just a moment what attaches over here to 7. All right, so let's continue to focus on our isolated peaks.
40:21
So if we look at 4 and we go with 4 we have 9J to 4, we have 7H to 4, we have 6D and E and maybe the better way to do it is 6DE to 4
40:45
and we have C to 4. Now, all right we have 4 and we're pretty sure
41:01
at this point we're pretty sure we have this alcohol, we're pretty sure that C is the alcohol, it's isolated. I think at this point we really can say all right here's our 8C, we're on 4. Now, 4 is going to be a very, very important linchpin
41:23
and I guess the thing that's confounding to me at this point is the fact that we've got 9J to 4 but that's not necessarily saying that that isolated methyl is directly attached to 4. We're running out of atoms so we're going to be good in a second. We have 7 to 4, we have 6D and E to 4,
41:44
they're probably all attached but watch what happens if I now pick up on another isolated one so I probably can start to infer things but let's continue with another one of our problem children, another one of our isolated ones. So let's go off of C. We wouldn't be seeing cross peaks
42:03
off of C if it were exchanging rapidly. If the alcohol on C were not residing on that proton, on that oxygen for several hundredths or a tenth of a second if it were exchanging rapidly, usually the less sterically congested alcohol the more
42:23
rapidly it exchanges so usually a primary alcohol exchanges rapidly, usually sometimes a second is secondary is slow, sometimes it's fast, usually a tertiary tends to be slower, just more steric congestion, harder for a molecule of water to get in and make it exchange.
42:41
Deacidify your chloroform, pass it through alumina, not your sample but just your chloroform through flame or furnace dried alumina, it's a good way to reduce exchange because you're getting rid of acid in the sample but here we happen to see beautiful, beautiful coupling off of C and so if you look at your cross peaks here and again we're going to use things
43:01
in this focused way, we have C9, C7, C6 and so you look at all of these cross peaks and that really is nice because remember you're going to be picking up your 2
43:23
and 3 bond coupling unless you have any intervening coupling through a double bond where sometimes you can pick up 4 bond coupling, remember sometimes like acetylenes we saw how weird the acetylene behaved, I mentioned it's a 2 bond, it's a 50 hertz coupling so you can get homo-propergolic coupling
43:42
but here this is pretty nice, we've already taken care of 2 of our bonds so look how valuable this is as a lynch pin, every cross peak now we get is going to help give us our connectivity so we have C4,
44:01
that's going to connect over to C6 because we're getting that cross peak here so we know that this is 3 bond, that's as far as we can go, we're really lucky at this point. Now we get this cross peak over to 7
44:20
and so again that's just sewing this whole molecule together, we have this cross peak to 9 and we said already here remember we had this fragment, we weren't able to place it but it could be redundant because we were running out of carbons, we have this cross peak over to 9 and so there's our C9H3J
44:44
and now this whole molecule is sewn up. So if you look at our structure right now, I'll just redraw this and I guess I'll draw it real small up here so here's the whole structure of our molecule,
45:15
we have our 7 and 8 over here, our 6 here, our ethyl group, our isolated methyl, our isolated hydroxy
45:24
and this set of cross peaks here has really been key. Now the good news is there are other isolated points in this molecule that can put things together so when I try to attack a problem I'm going to start with the isolated ones, obviously eventually you want
45:42
to go back and sort of check yourself and see that everything is consistent but let's take another point of attack here. So let me just run down this track of 9J and see who 9J is crossing with. So 9J is crossing with 7, 9J is crossing with 6
46:07
and 9J is crossing with 4 and so you look at that and you say, oh, okay, that's giving me the same information. See the HMBC is really screaming out at me
46:21
through the isolated ones. It was a gif that we got HC. HC might not have coupled. It's an alcohol. HC might have been exchanging rapidly if there was some acid in our sample. It was a gif that we got it. It just gave us the whole problem but we get that same gif right off of carbon 9 because 9
46:42
with J is giving us 7 so we're crossing over to here and again remember your hydrogen, you have one bond from hydrogen J to carbon, one bond from carbon 9 to carbon 4 and so any other cross peak now has to be attached directly to carbon 4.
47:01
So we see a cross to 7, we see a cross to 6 and we see a cross to 4 that's our two bond coupling so that's giving us the information. I'll take one more isolated carbon just for the heck of it. It's not really completely isolated but it's 10G
47:21
and so if we go off of 10G, let's see, we have here 10G with 3 and 10G with 8 over there and so if you'll look at that, that's also giving us some information. We had the 10G to 3 biallelic coupling by the proton NMR
47:45
so that information was nothing new but you don't get a little coupling right through here so this information here of 10G to 8 is actually useful. Had we not been able to place, had we not been able
48:01
to place 8 directly on 2? Had we not picked up that a little coupling or had things been more complicated or confusing or had we only at that point had this HMBC cross peak between 2 and 7 and 8? Remember how I said 2 is crossing with both HF and HHH
48:25
and we couldn't be sure about that? If I wasn't certain at that point then we could have gotten that later on from this cross peak so there's multiple pieces of evidence all pointing in this direction.
48:42
Anyway, this is how I view HMBC as being incredibly powerful for putting pieces together. I want to show you a couple of last things that are sort of common features that are kind of cool.
49:04
They're also germane to some of the upcoming homework so it's worth actually keeping in mind. One thing that's very cool is okay if you have 2 CH3s on a methyl so remember how I said you can get these vampire
49:23
bites, you can get your 1 bond couplings but if you're going ahead and if you have no stereocenter, in other words if these methyls are either not diastereotopic or they're coincident then your HMBC even if you see these vampire type, bite type of cross peaks here,
49:43
even if you see, so here's your methyl and here's your methyl in the C13, here's your methyl in the H1, even if you see those types of cross peaks this is a real HMBC cross peak here
50:02
because the 2 methyls are crossing with each other so that's kind of cool. Well in this particular cartoon example they're magnetically equivalent, if they were magnetically inequivalent if they were diastereotopic then you'd still see HMBC peaks. By the way, the tough thing
50:21
about HMBC is you're not guaranteed to get a cross peak because you're particularly on 3 bond so 2 bond is weird because your 2 bond Js end up being all over the map they'll usually show up but not always. 3 bond is weird in a different way, it's weird by a dihedral angle, by a carplus relationship.
50:44
So if you're hydrogen, so if you have a dihedral angle that's defined as HCCC like so, if you have good overlap between this hydrogen and this carbon,
51:01
a good geometrical overlap meaning an antiperiplanar or a synperiplanar relationship you typically get a big J. It shows up in the HMBC but if these 2 are at close to 90 degree angles you're not going to get a cross peak typically and so that's the third thing
51:20
about HMBC that's really confusing. The beauty of methyl groups is with a methyl group no matter what you're always going to have, at least you're always going to have protons at a good dihedral angle. So methyl groups, isolated methyl groups in general end up being extremely valuable for HMBC but we also see here
51:44
that the isolated methyl groups are usually the problem children because you don't know where to put them and fortunately they are talkative problem children because they will talk to you in the HMBC. So that's a very useful one. A couple of other things just to keep in mind,
52:04
carbonyls if you've got like this, if you've got something like this, whatever, the problem is you don't know if it's 2 bond coupling or 3 bond coupling but if you can build up your pieces and say oh well I've got both of them then you can figure
52:20
out alright that carbonyl is attached to one and the other. Let's see, I guess I don't know if this comes up. We already saw an ester, some people have said to me oh I didn't know that you could get coupling through heteroatoms but yeah coupling occurs through carbon oxygen bonds, through carbon carbon bonds, through carbon nitrogen bonds.
52:40
So for example and in a mid here like so, again your carbonyl can be a real lynch pin. You can get all different protons coupling with that carbonyl so they can end up telling you lots
53:01
and lots of information for putting the pieces together. But what's nice is carbonyls often isolate one spin system from another spin system and it's those isolated spin systems, what I'm calling fragments, that are so hard to put together and that's why HMBC is so valuable as a focus tool.
53:22
Alright this should give you the skills to attack the problems not for this week's homework set but for the next week's homework set where the molecules are going to get more complicated and where you're going to be more dependent on being able to systematically put your pieces together.
53:41
Yeah? You couldn't tell, great question, James is asking could we tell the stereochemistry of our center? No, the racemate would have the exact same spectra as the one enantiomer or the other enantiomer.
54:02
The only way to differentiate them would be to add something chiral to make a Mosher ester here for example with a chiral derivatizing agent or to use what's called a chiral shift reagent which I haven't planned a section on for our course but it's basically a Lewis acid that's chiral
54:21
that will interact with the two enantiomers differently. So you need chirality to distinguish chirality.