Millimeter-wave rotary-wing (RW) unmanned aerial vehicle (UAV) air-to-ground (A2G) links face unpredictable Doppler effect arising from the inevitable wobbling of RW UAV. Moreover, the time-varying channel characteristics during transmission lead to inaccurate channel estimation, which in turn results in the deteriorated bit error probability performance of the UAV A2G link. This paper studies the impact of mechanical wobbling on the Doppler effect of the millimeter-wave wireless channel between a hovering RW UAV and a ground node. Our contributions of this paper lie in: i) modeling the wobbling process of a hovering RW UAV; ii) developing an analytical model to derive the channel temporal autocorrelation function (ACF) for the millimeter-wave RW UAV A2G link in a closed-form expression; and iii) investigating how RW UAV wobbling impacts the Doppler effect on the millimeter-wave RW UAV A2G link. Numerical results show that different RW UAV wobbling patterns impact the amplitude and the frequency of ACF oscillation in the millimeter-wave RW UAV A2G link. For UAV wobbling, the channel temporal ACF decreases quickly and the impact of the Doppler effect is significant on the millimeter-wave A2G link.