PM2.5, generated via both direct emission and secondary formation, can have varying environmental impacts due to different physical and chemical properties of its components. However, traditional methods to quantify different PM2.5 components are often based on online or offline observations and numerical models, which are generally high economic cost- or labor-intensive. In this study, we develop a new method, named Multi-Tracer Estimation Algorithm (MTEA), to identify the primary and secondary components from routine observation of PM2.5. By comparing with long-term and short-term measurements of aerosol chemical components in China and the United States, it is proven that MTEA can successfully capture the magnitude and variation of the primary PM2.5 (PPM) and secondary PM2.5 (SPM). Applying MTEA to the China National Air Quality Network, we find that (1) SPM accounted for 63.5 % of the PM2.5 in cities in southern China on average during 2014–2018, while the proportion dropped to 57.1 % in the north of China, and at the same time the secondary proportion in regional background regions was ∼ 19 % higher than that in populous regions; (2) the summertime secondary PM2.5 proportion presented a slight but consistent increasing trend (from 58.5 % to 59.2 %) in most populous cities, mainly because of the recent increase in O3 pollution in China; (3) the secondary PM2.5 proportion in Beijing significantly increased by 34 % during the COVID-19 lockdown, which might be the main reason for the observed unexpected PM pollution in this special period; and finally, (4) SPM and O3 showed similar positive correlations in the Beijing-Tianjin-Hebei (BTH) and Yangtze River Delta (YRD) regions, but the correlations between total PM2.5 and O3 in these two regions, as determined from PPM levels, were quite different. In general, MTEA is a promising tool for efficiently estimating PPM and SPM, and has huge potential for future PM mitigation.