When we study the Berry phase, we're dealing with the time dependent parameters with manifold M and the adiabatic transport of the state, which for N×N-dimensional hamiltonian defines the mapping M→CPN−1,
where
CPN−1 is the space of N complex unit vectors defined up to the phase. Suppose mapping
M→S2N−1
where
S2N−1 is the space of complex unit vectors with definite phase. During adiabatic transport this mapping is possible if the phase can be globally defined. I want to clarify if there exists a lifting between these mappings
(1),(2):
f:M→CPN−1 to ˜f:M→S2N−1
Suppose
M=Sn,n>1. Then in order to check whether the lifting is possible, have I to compare the homotopy groups
πi(CPN−1) with πi(S2N−1),i=1,...,n,
or only
πn(CPN−1) with πn(S2N−1)?