Embodied carbon in trade has attracted the attention of researchers, developers, and industry leaders worldwide, as it has become a considerable source of trade conflict. Because China is the world’s leading carbon-emitting country, it is important to study the change trends of embodied CO_2 in Chinese trade to identify key influencing factors. This paper investigates the trade embodied CO_2 in China, with special focus on the production process and materials of each industrial sector from 1990 to 2013. This study was conducted by using an input-output model, which analyzes the specific causes of periodic waves. Our findings suggest that the machine manufacturing sector produced the largest share of emissions, whereas the metal and non-metal sector has the highest emission intensity. Moreover, the total emission quantities in trade increased from 612 to 3331 million tons in 1990 to 2013, a period characterized by decentralized trends in emissions exported to trade partners. We also employed the log-mean Divisia index(LMDI) model to decompose the carbon emissions changes into export structure, export scale, energy structure, and energy intensity, and estimate the influence of these factors at a number of time points. We found that energy intensity curbs emissions increase, and that export scale expansion is the strongest driving force. Embodied carbon in trade has attracted the attention of researchers, developers, and industry leaders worldwide, as it has become a substantial source of trade conflict. China is the world’s leading carbon-emitting country, it is important to study the change of trends CO_2 in Chinese trade to identify key influencing factors. This paper investigates the trade embodied CO_2 in China, with special focus on the production process and materials of each industrial sector from 1990 to 2013. This study was conducted by using an input-output model, which analyzes the specific causes of periodic waves. Our findings suggest that the machine manufacturing sector produced the largest share of emissions, but the metal and non-metal sector has the highest emission intensity. Moreover, the total emission emission in trade increased from 612 to 3331 million tons in 1990 to 2013, a period characterized by decentralized trends in emissions exported to trade partners. We also emp loyed the log-mean Divisia index (LMDI) model to decompose the carbon emissions changes into export structure, export scale, energy structure, and energy intensity, and estimate the influence of these factors at a number of time points. We found that energy intensity curbs emissions increase, and that export scale expansion is the strongest driving force.