Computational thinking (CT) is necessary for Science, Technology, Engineering, and Mathematics (STEM) literacy, but it can be difficult for many students to develop and it is challenging to integrate into science curricula. Here, we present a five-session curriculum where sixth-grade students programmed a Liquid Handling Robot (LHR) to conduct a science experiment while engaging in CT. We used a mixed-methods approach to assess how the curricular integration of robotics and science experimentation advances students' CT skills and perceptions of computation in science. We identified growth in CT skills, specifically regarding Algorithmic Thinking. Students identified as key advantages of this approach the increased precision in experimental procedures, time-efficiency, and easier debugging. This course provides a proof of concept curriculum on how the implications for teaching and learning of CT can be assessed, and how CT and robotics can be brought to science classrooms, especially for chemistry and biology.