In your book, you apply mean field theory to the Ising Model. However, in the problem set (question 5.3), we are asked to apply mean field theory to the Heisenberg Model. Is that what was expected for the question, or did you have the Ising Model instead in mind?
I can't see how we would be able to obtain the partition function without assuming strong anisotropy (ie. Ising Model) because it involves the dot product of two spins, which is not trivial...
« Back to index