Abstract. Solar coronal mass ejections can accelerate charged particles, mostly protons, to high energies, causing Solar Particle Events (SPEs). Such energetic particles can precipitate upon the Earth’s atmosphere, mostly in polar regions because of the geomagnetic shielding. Here, SPE induced chlorine activation due to ion-chemistry can occur and the activated chlorine depletes ozone in the polar middle atmosphere. We use a state of the art 1D stacked-box model called Exoplanetary Terrestrial Ion Chemistry (ExoTIC), of atmospheric ion and neutral composition to investigate such events in the Northern Hemisphere (NH). Measurement data from the Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) on Environmental Satellite (ENVISAT) were used to evaluate the model results using the Halloween SPE in late October 2003, a well-known large event, as a test field. Sensitivity tests were carried out for different model settings with a focus on the chlorine species of Hypochlorous acid (HOCl) and Chlorine Nitrate (ClONO2) as well as ozone and odd oxides of nitrogen (NOy). The model studies were carried out in the northern hemisphere for a high latitude of 67.5° N, inside the polar cap. Comparison of the simulated effects against MIPAS observations for the Halloween SPE revealed a rather good temporal and spatial agreement for HOCl, ozone and NOy. For ClONO2, a good spatial agreement was found. The best model setting was the one with full ion-chemistry where oxygen atom in the excited state, O(1D) was set to photo-chemical equilibrium. HOCl and ozone changes are very well reproduced by the model, specially for night-time. HOCl was found to be the main active chlorine species under night-time conditions resulting in an increase of more than 0.2 ppbv. Further, ClONO2 enhancements of 0.2–0.3 ppbv have been observed both during daytime and night-time. In a nutshell, the most appropriate model setting delivers satisfying result, i.e. the model can be considered to be positively validated. Model settings that compared best with MIPAS observations were applied to an extreme solar event in 775 A.D., presumably a once in a 1000 year event. With the model applied to this scenario, assessment can be made what is to be expected at worst for effects of a SPE on the middle atmosphere. Here, a systematic analysis comparing the impact of the Halloween SPE and the extreme event on the Earth’s middle atmosphere is presented. As seen from the model simulations, both events were able to perturb the polar stratosphere and mesosphere, with a high production of NOy and odd oxides of hydrogen (HOx). Longer lasting and stronger stratospheric ozone loss was also seen for the extreme event. Qualitative difference between the two events and a long lasting impact on HOCl and hydrochrolic acid (HCl) for the extreme event was found. Chlorine ion-chemistry contributed to a stratospheric ozone loss of 2.4 % during daytime and 10 % during night-time during the Halloween SPE as seen with time dependent ionisation rates applied to the model. Furthermore, while comparing the two events just for the event day, an ozone loss of 10 % and 20 % was found during the Halloween SPE and the extreme event respectively which was due to the impact of chlorine ion-chemistry.