Random Copolymer: Gaussian Variational Approach

(Received 18 September 1996, received in final form 4 December 1996, accepted 10 December 1996)

Abstract
We study the phase transitions of a random copolymer chain with quenched disorder. We calculate the average over the quenched disorder in replica space and apply a Gaussian variational approach based on a generic quadratic trial Hamiltonian in terms of the correlation functions of monomer Fourier coordinates. This has the advantage that it allows us to incorporate fluctuations of the density, determined self-consistently, and to study collapse, phase separation transitions and the onset of the freezing transition within the same mean field theory. The effective free energy of the system is derived analytically and analyzed numerically in the one-step Parisi scheme. Such quantities as the radius of gyration, end-to-end distance or the average value of the overlap between different replicas are treated as observables and evaluated by introducing appropriate external fields to the Hamiltonian. As a result we obtain the phase diagram in terms of model parameters, scaling for the freezing transition and the dependence of correlation functions on the chain index.