<p>The innovation and diffusion of renewable energy technologies are critical to advancing the energy transition. This study investigates how knowledge acquisition through spillovers from renewable and non-renewable patented technologies affects the subsequent diffusion of these technologies. Using a dataset comprising 12,966 renewable energy patent families and 24,067 non-renewable (fossil fuel-based) patents owned by firms between 1990 and 2010, we analyse backward citations as indicators of knowledge spillovers and forward citations to trace knowledge diffusion. We classify spillovers into three types: clean spillovers, dirty spillovers, and external spillovers (originating from non-energy fields). Our results show that, for renewable technologies, clean spillovers have a positive and significant effect on diffusion, while dirty spillovers hinder it, indicating a clean knowledge path dependence. In contrast, for non-renewable technologies, dirty spillovers significantly promote diffusion toward other non-renewable technologies, confirming a dirty knowledge path dependence. Green absorptive capacity does not moderate spillovers in the case of renewables, but it does reduce the influence of non-renewable spillovers on the diffusion of dirty technologies. These findings have important implications for innovation policy and knowledge transfer mechanisms aimed at accelerating the energy transition.</p>