[StS]

I know when dealing with one multiplet in H NMR just use pascal's triangle for relative intensities. But for a triplet of triplet are the relative intensities sums of each other and therefore would be: 1:3:1:1:4:1:1:3:1 or products and be: 1:2:1:1:4:1:1:2:1?

[Vidya]

in this case you should use J-J coupling

[Dan]

I'm not really sure where you got those numbers from.

I think about it like this:

Take a 1:2:1 triplet, and scale up to 4:8:4 (still the same ratio).

Now split each of those 3 peaks into a 1:2:1 triplet:

4 becomes 1:2:1
8 becomes 2:4:2
4 becomes 1:2:1

overall, the tt is 9 peaks with ratio 1:2:1:2:4:2:1:2:1

[Dan]

I see now what Uma was getting at (I think). The pattern I posted is assuming that one of the coupling constants is much larger than the other, specifically that J₁ > 2J₂, which means that there is no overlap of the individual triplet patterns.

If J₁ = 2J₂, then some peaks will overlap (you might need to draw a tree diagram to see it):

1:2:1:2:4:2:1:2:1

The red peaks are coincident, giving an apparent septet:

1:2:3:4:3:2:1

If J₂ < J₁ < 2J₂

Then the highlighted peaks will swap (again tree diagram will help)

1:2:1:2:4:2:1:2:1

becomes

1:2:2:1:4:1:2:2:1

[StS]

I'm not really sure where you got those numbers from.

I think about it like this:

Take a 1:2:1 triplet, and scale up to 4:8:4 (still the same ratio).

Now split each of those 3 peaks into a 1:2:1 triplet:

4 becomes 1:2:1
8 becomes 2:4:2
5 becomes 1:2:1

overall, the tt is 9 peaks with ratio 1:2:1:2:4:2:1:2:1

Sorry I was assuming non-overlapping peaks, so this is good. What I was trying to figure out was if the 1 of the first triplet was split into another triplet would I sum 1 to each 1:2:1 or multiply it to get the correct intensities. I see now that you multiplied them, thanks (I messed that up in my original post).