Phase field models have enjoyed a great deal of attention, due to their flexibility in capturing the essential thermodynamics and symmetry of physical phenomena, while simultaneously allowing for the simulation of extremely complex microstructure formation. Over the past 10 years, the ability to model solidification, grains, and stress (in addition to solid-solid transitions) has allowed us to begin to model some of the more technologically important processing techniques. As the phase field approach starts from the solid foundation of thermodynamics, and includes only a a modest amount of phenomenology, the predictive power of this approach is well demonstrated. This talk will examine the application of the phase field method to materials containing grain boundaries, and explore some an unexpected puzzle in pattern formation.