High levels of secondary air pollutants during COVID-19 in China have aroused great concern. In Beijing, measured daily mean peroxyacetyl nitrate (PAN) concentrations reached 4 ppb over the lockdown period (24 January to 15 February), whose averages were 2-3 times that before lockdown (1-23 January). The lockdown PAN levels also reached a high historical record based on our long-term measurements (2016-2019). Unlike ozone and PM 2.5 , PAN formation depends on less complex photochemistry between NO x and volatile organic compounds (VOCs), providing a novel approach to investigate the wintertime photochemistry during COVID-19. The GEOS-Chem simulations suggest a markedly enhanced photochemistry by a factor of 2 during the lockdown. Change of meteorology featuring with anomalous wind convergence under higher temperatures is the main reason for enhanced photochemical formation of PAN, while chemically nonlinear feedbacks also play a role. Our results suggest implementing targeted VOC emission controls in the context of increasing photochemical pollution over this complex polluted region. Plain Language Summary Outbreaks of the COVID-19 pandemic caused immediate implementation of lockdown policy in China, which drastically decreased emissions of primary air pollutants. Peroxyacetyl nitrate (PAN), as an important photochemical product, is controlled by reactions between NO x and volatile organic compounds (VOCs) that were reduced substantially due to the lockdown. However, observed PAN levels in Beijing during the lockdown were markedly enhanced and were even much higher than the concentrations during the same periods in 2016-2019. Modeling results prove that this increase in PAN is driven by enhanced photochemistry, resulting from anomalous wind convergence under higher temperature and enhanced radical level in response to NO x reduction. Our results suggest the necessity of reducing VOC emissions in controlling photochemical pollution even in the wintertime over China.