Particulate chloride is an important component of fine particulate matter in marine air masses. Recent field studies also report elevated concentrations of gas-phase reactive chlorine species and particulate chloride related to anthropogenic activities. This work focuses on particulate chloride detection and quantification issues observed for some quadrupole aerosol chemical speciation monitors (Q-ACSM), which are designed for long-term measurement of ambient aerosol composition. The ACSM reports particle concentrations based on the difference between measurements of ambient air (sample mode) and particle-free ambient air (filter mode). For our long-term campaign in Krakow, Poland, the Q-ACSM reports apparent negative total chloride concentration for most of the campaign when analyzed with the default fragmentation table. This is the result of the difference signal from m/z 35 (35Cl+) being negative which dominates over the positive difference signal from m/z 36 (H35Cl+). Highly time-resolved experiments with NH4Cl, NaCl and KCl particles show that the signal response of m/z 35 is non-ideal, where the signal builds up and decreases slowly for all three salts, leading to a negative difference measurement. In contrast, the m/z 36 signal exhibits a near step-change response for NH4Cl during sampling and filter period, resulting in a positive difference signal. The response of m/z 36 for NaCl and KCl is not as prompt as for NH4Cl but still fast enough to have a positive difference signal. Furthermore, it is shown that this behavior is mostly temperature-independent. Based on these observations, this work presents an approach to correct the chloride concentration time series by adapting the standard fragmentation table coupled with a calibration of NH4Cl to obtain a relative ionization efficiency (RIE) based on the signal at m/z 36 (H35Cl+). This correction can be applied for measurements in environments where chloride is dominated by NH4Cl. Caution should be exercised when other chloride salts dominate the ambient aerosol. This repository contains the data corresponding to each figure presented in the published main text.