some complex signals (especially diastereotopic protons with lots of different neighbors or signals corresponding to homotopically different protons that happen to overlap) are really just too complex, and we label them multiplets and move on.
For well separated signals, it is most accurate to tease out the various J values and report them. But finding all the J values for a dddd signal is not always straightforward. I use the 1994 paper to match up my signal with the predicted signals in the paper to determine exactly what the splitting is (dddd, dt, etc). Then I'll go to the 2002 paper to actually calculate the J values.
The flow chart in Chart 1 is really what you need. Convert all your ppms to Hz and list them largest to smallest. Notation like {1 to 2} refers to the signal peak numbered from left to right. For each 'd' in the splitting pattern, you get d2 peaks. A d is 2 peaks, dd is 4, ddd is 8, dddd is 16, etc... However, due to equivalent J values, etc, some of the peaks overlap within a signal. So a dddd which should have 16 peaks, may only show 10 peaks - of various heights. The taller peaks correspond to accumulated smaller peaks... but all 16 peaks should be there (see Fig 2 in the 2002 paper). So find out where all your d2 peaks are and number them 1 - d2. Then use the Flow chart to find J1, J2, J3, J4, ...
ChemBuddy | 
Chem Reddit | 
Topic: NMR Multiplicities (Read 8020 times)