The temperature dependent geometric phase

Referee: [H₂⁺ demonstration] The central construction treats the system-environment interaction as a BO separation in which the environment stays in a thermal mixed state ρ_env = exp(−H_env/kT)/Z. Standard Berry phase requires unitary adiabatic evolution of a pure state with a well-defined instantaneous eigenstate. The manuscript must show explicitly (in the H₂⁺ section) that the connection extracted after tracing over the environment remains purely geometric—i.e., path-dependent and independent of traversal speed—rather than acquiring dynamical or dissipative contributions from the mixed-state evolution.

Authors: We thank the referee for this important clarification. In the revised manuscript we have expanded the H₂⁺ section with an explicit calculation of the effective connection after tracing out the environment. The geometric phase is obtained as the line integral of the thermally averaged Abelian connection A_eff = Tr_env[ρ_env(T) A], where A is the microscopic Berry connection. We separate the total phase acquired along a closed path into dynamical and geometric contributions, showing that the geometric part depends only on the path in parameter space and remains independent of traversal speed in the adiabatic limit. While the environmental state is mixed, the effective dynamics under the BO approximation isolates the geometric contribution without dissipative terms. revision: yes

Referee: [Gauge potential derivation] The claim that the Abelian gauge potential is induced by the BO approximation and is temperature-dependent appears to rest on inserting the thermal distribution into the effective Hamiltonian. The derivation should be checked for circularity: if the temperature dependence is introduced by averaging over the environment's thermal state rather than emerging from the adiabatic connection itself, the result reduces to a redefinition rather than a new geometric effect. A concrete counter-check would be to compute the phase for a closed unitary path and verify that it matches the standard Berry phase when T→0.

Authors: We have clarified the origin of the temperature dependence in the revised text. The Abelian gauge potential arises from the parametric dependence of the fast environmental degrees of freedom on the slow system coordinate within the BO separation; temperature enters because the environmental state is thermal. This is not a simple redefinition but follows from averaging the microscopic connection over the thermal ensemble. We now include an explicit counter-check: as T → 0 the thermal state reduces to the ground-state projector and the extracted phase recovers the standard Berry phase for the pure-state unitary evolution, confirming that the geometric character is preserved. revision: yes

Referee: [BO-like approximation] The adiabatic procedure between system and environment is asserted to be analogous to BO, yet the environment is a continuum of modes in thermal equilibrium. The manuscript should specify the timescale separation (system slow, environment fast) and derive the effective gauge potential without assuming the environment remains unperturbed. If the back-action on the environment is neglected, the resulting phase may not be gauge-invariant under local unitary transformations on the system alone.

Authors: We agree that the timescale separation requires explicit statement. The revised manuscript now specifies that the environment thermalizes on a timescale much shorter than the adiabatic evolution of the system, allowing it to remain in instantaneous thermal equilibrium. The effective gauge potential is obtained by first solving the fast environmental problem and then averaging, which incorporates the leading back-action through the interaction term. We have added a short argument showing that the resulting Abelian connection transforms correctly under local unitary transformations on the system, so that the holonomy (geometric phase) for any closed path remains invariant. revision: partial