There is growing recognition in the education community that the problem-solving practices that comprise computational thinking (CT) are a fundamental component of both life and work in the twenty-first century. Historically, opportunities to learn CT have been confined to computer science (CS) and elective courses that lack racial, ethnic, and gender diversity. To combat this inequity, a number of scholars have proposed integrating CT practices into core curriculum--especially science, technology, engineering, and math curriculum. Successfully achieving the goal of integrated CT, however, depends on frameworks to guide integration, professional development for teachers, exemplars of successful integrations, and identifications of the barriers teachers encounter. Research pertaining to each of these areas is in its infancy. This study addresses these needs through a collective case study of 10 secondary science teachers' implementations of a novel, process-based, unplugged approach to CT/science integration and the factors that supported or hindered their CT/science integration efforts. The results of this work reveal that: (1) an unplugged and process-based approach to CT/science integration shows promise as a vehicle for infusing CT into diverse science classrooms; (2) educators' teaching context exerts a strong influence on their CT-integration efforts and persistence; and (3) special attention is needed to support teachers in their CT/science integrations including algorithm creation. This study also demonstrates the utility of the Fraillon et al.'s CT framework as a guide for CT/science integration efforts and sheds light on the unique affordances of unplugged strategies for implementing CT-integrated science curricula.