Abstract
In the context of the low-carbon transition and sustainable development, this paper investigates risk spillovers between carbon and energy chemical markets by adopting a frequency-domain multi-layer network approach. Spillovers are decomposed into short-, medium-, and long-term frequency layers to distinguish risk spillovers across different time scales. Intra-layer weighted directed networks are constructed for each layer, and inter-layer connections are established between adjacent layers to form a multi-layer network. Change point identification is then applied to divide the sample into six phases. The empirical results are summarized as follows: First, carbon markets are predominantly net risk receivers, while energy chemical markets are predominantly net risk transmitters, although exceptions exist in both groups. Second, spillovers are dominated by the short-term frequency layer and diminish as the time scale lengthens. Third, the structural importance of carbon markets is generally lower and uneven across markets and frequency layers, whereas energy chemical markets maintain relatively stable cross-frequency-layer connections. Finally, the network positions of carbon markets evolve across phases, with key markets rotating over time, while energy chemical markets remain structurally dominant. These findings provide a frequency-domain multi-layer network perspective for understanding risk spillovers between carbon and energy chemical markets and offer implications for collaborative risk regulation, thereby contributing to financial stability and sustainable market development during the green transition.