925-456-1020 info@acornnmr.com

Assignment strategy

Assignment strategy

Strategy for NMR Assignments of Codeine

An analysis such as described below is a common component of characterization of reference materials and new drug applications to FDA.  Typically, a short report is written describing the work performed and listing the assignments, with a brief explanation of how the assignments were made.  All spectra are returned to the customer with the report, including expanded plots as needed to show all experimental detail.

Starting with the 1H spectrum, some features are immediately obvious (2 aromatics, 2 methyl singlets).  The rest of the spectrum consists of numerous multiplets that integrate to 1 H each.  It is not clear how many of these might be non-equivalent CH2s.

One of the first things we need to know is how many carbons are there, and how many of each multiplicity – quaternary, CH, CH2 and CH3.  So the next step, assuming there is enough material, is a 1D 13C spectrum, which shows we have 18 Cs.  (There are 2 peaks at ~43 ppm).

It is helpful to have an unambiguous label for each carbon.  We find it convenient to number the carbons from 1 to n starting downfield.

There are 2 options for determining multiplicity – DEPT and HMQC (or the combined experiment, DEPT-HMQC).   DEPT will unambiguously determine multiplicities, but HMQC has the advantage of also "connecting" 1H and 13C peaks, so information from each can be used to assign the other.  For example, methyl peaks are often obvious in a 1H spectrum, allowing us to identify methyls in the 13C spectrum.  The HMQC also allows us to pair up geminal CH2 1H peaks, which show up in the HMQC as 2 peaks at the position of a single 13C.  3 CH2s are easily identified in the 2D plot below.

At this point, we can create the following table:

Peak

ppm 

  Mult

Peak

ppm 

  Mult

 1  

146.38  

q  

10  

 66.43  

CH  

 2  

142.23  

q

11  

 58.92  

CH  

 3  

133.43  

CH  

12  

 56.40  

CH3  

 4  

131.13  

q

13  

 46.47  

CH2  

 5  

128.30  

CH  

14  

 43.12  

CH3  

 6  

127.30  

q

15  

 42.99  

q

 7  

119.58  

CH  

16  

 40.82  

CH  

 8  

113.03  

CH  

17  

 35.85  

CH2  

 9  

 91.39  

CH  

18  

 20.46  

CH2  

 

From the 13C and DEPT and/or HMQC, we determine the structure contains:

8 CHs
3 CH2s
2 CH3s
and the remaining 5 must be quaternary Cs.

This gives a total of 20 Hs.  There is also an OH.  But this doesn’t match the integration in the 1H spectrum above.  The peak at 1.87 ppm, with integral of 2.49, is actually just 1 H, overlapping with H2O.  Compare spectra at different concentrations, below.  At lower concentration, water is shifted upfield approximately .5 ppm.

 

From the HMQC, we connect each 1H peak to its 13C peak, so we now have numerical labels for all Hs.

 

13C (ppm) 

 

1H (ppm)

 1  

146.38  

q  

 

 2  

142.23  

q

 

 3  

133.43  

CH  

5.71

 4  

131.13  

q

 

 5  

128.30  

CH  

5.29

 6  

127.30  

q

 

 7  

119.58  

CH  

6.57

 8  

113.03  

CH  

6.66

 9  

 91.39  

CH  

4.89

10  

 66.43  

CH  

4.18

11  

 58.92  

CH  

3.35

12  

 56.40  

CH3  

3.84

13  

 46.47  

CH2  

2.59, 2.40

14  

 43.12  

CH3  

2.44

15  

 42.99  

q

 

16  

 40.82  

CH  

2.67

17  

 35.85  

CH2  

2.06, 1.88

18  

 20.46  

CH2  

3.04, 2.30

  

  

OH

2.99

 

Assignments

The next step is examination of 1H and 13C spectra to identify those peaks whose assignment is obvious based on chemical shift.

We can identify one OMe (3.84 ppm) and one NMe (2.44 ppm) in the proton spectrum, and can connect them to carbons 56.4 ppm (#12) and 43.1 ppm (#14), respectively, from the HMQC spectrum.

The aromatic Hs are obvious as the 2 doublets at ~ 6.6 ppm with a mutual coupling of 8 Hz, but it is not clear which is which.  The corresponding aromatic CH carbons are 119.6 (#7) and 113.0 ppm (#8).  Both are either ortho or para to O substituents, so are upfield.

The olefinic Hs are at 5.71 and 5.29 ppm, with corresponding Cs at 133.4 and 128.3 ppm (#3 and 5), respectively, but again it is not clear which is which.

Carbon #9 at 91.4 ppm (H at 4.89 ppm) and carbon #10 at 66.4 ppm (H at 4.18 ppm) can be assigned to the epoxide and hydroxyl carbons, respectively.

The most downfield carbons (#1 and 2 at 146.4 and 142.2 ppm) are expected to be the O-substituted aromatics, but it is not clear which is which.

The 2 remaining quaternary aromatics must be #4 and 6 (127.3 and 131.1 ppm), but it is not obvious which is which.

The only aliphatic quaternary C is #15 at 43.0 ppm, so it is assigned by process of elimination.

Some additional assignments can be made from the COSY and NOESY spectra.  

A COSY cross-peak from the OH (2.9 ppm) to 4.18 ppm confirms this as the CH bearing OH (#10).  

A NOESY cross-peak from OMe to the aromatic H at 6.66 ppm identifies the latter as (#8) ortho to the methoxy, distinguishing between #7 and 8.

A NOESY cross-peak from H-3 to H-10 identifies the H at 5.71 ppm as adjacent to #10 and resolves the uncertainty of #3 and 5.

NOESY crosspeaks from H-7 to H-18 and H-18′ allows #18 to be assigned.

Assignments so far are summarized in the structure below:

The HMBC (multiple-bond C-H correlation) spectrum will allow the remaining assignments to be made.

The methoxy protons have a strong 3-bond coupling, in the HMBC spectrum, to the aromatic carbon bearing OMe.  This allows assignment of C-1 to carbon bearing OMe and, by elimination, C-2 is the C bearing the epoxide O. 

The HMBC is optimized for 8 Hz couplings, typical of aromatic 3-bond couplings.  The HMBC shows correlations from H-8 to C-1 and C-6, and from H-7 to C-2 and C-4.  This confirms assignment of C-1 and C-2, and allows assignment of C-4 and C-6.

The remaining ambiguities are distinguishing CHs #11 and 16, and CH2s #13 and 17, plus stereo-specific assignments for Hs 13, 17 and 18.

An HMBC peak from the N-Me protons (#14) to C-11 (3-bond) resolves the ambiguity between #11 and #16.

An HMBC peak from H-9 to C-17 (3-bond) resolves the ambiguity between #17 and 13.

The resulting assignments are:

The NOESY spectrum provides stereo-specific assignments of the methylene Hs.

The NOESY spectrum shows an NOE between H-5 and H-18′.

H-5 is shown in yellow, H-18′ is green.

If you have the Chime plug-in, you can view this interactively in 3D.

The NOESY spectrum shows an NOE between H-18 and H-13.

H-18 is shown in yellow, H-13 is green.

The NOESY spectrum shows an NOE between H-16 and H-17.

H-16 is shown in yellow, H-17 is green.

 

 

The final step is to identify all peaks observed in the HMBC spectrum, looking for any inconsistencies in the assignments.  Any peak that would require an implausible coupling (too many bonds) suggests that an error has been made in the assignments.

C H # bonds C H # bonds
1 8 3 11 18, 18′ 2
1 9 3 11 13 3
2 8 2 11 14 3
2 7 3 13 17 2
2 12 3 14 and/or 15 11 3
3 9 3 14 and/or 15 13 3
4 7 3 15 17 2
4 18, 18′ 3 16 18 3
4 17 3 17 9 3
4 9 3 17 13 2
6 8 3 18 7 3
6 18, 18′ 2
7 18, 18′ 3

 

 


Last updated: 06/14/2010