We present a new nonlinear algorithm for the efficient and accurate solutio
n of isothermal and nonisothermal phase-change problems. The method correct
ly evolves latent heat release in isothermal and nonisothermal phase change
, and more important, it provides a means for the efficient and accurate co
upling between temperature and concentration fields in multispecies nonisot
hermal phase change. Newton-like superlinear convergence is achieved in the
global nonlinear iteration, without the complexity of forming or inverting
the Jacobian matrix. This "Jacobian-free" method is a combination of an ou
ter Newton-based iteration and an inner conjugate gradient-like (Krylov) it
eration. The effects of the Jacobian are probed only through approximate ma
trix-vector products required in the conjugate gradient-like iteration.