Purpose <p>The performance of five different low-cost particulate matter sensor models has been characterized with monodisperse DEHS particles by comparing particle number and mass concentrations in different size fractions to the results of an optical particle counter (Palas welas) used as reference.</p> Methods <p>The five investigated sensor models are the SDS011 (NovaFitness), SPS30 (Sensirion), NextPM (Groupe Tera), OPC-R2 and OPC-N3 (both Alphasense). To obtain information on the intra-specimen variability, five specimens of each sensor model were tested. Three different parameters were investigated, namely the number and mass-based detection efficiency between 0.15 and 5 μm for different particle concentration levels, accuracy of size fractionation, and dynamic response to rapid concentration changes.</p> Results <p>NextPM and SDS011 were less suitable for detecting DEHS particles than the other models. With the NextPM sensor, the measured number concentrations sometimes deviated by more than a factor of 10 from the reference. The other sensor models have their own advantages and disadvantages, depending on the purpose of the measurement. Compared to the OPC-N3 and OPC-R2, the SPS30 for instance is capable of detecting higher number concentrations, but only with a much lower size resolution than the OPC-N3 or OPC-R2.</p> Conclusions <p>Overall, the experiments have shown that the OPC-R2 and OPC-N3 sensors are the most suited ones for laboratory applications with DEHS-particles at concentrations of &lt; 1500 P/cm³. They have a comparably size-independent counting efficiency for d_P ≥ 0.5 μm and are able to accurately measure particle number size distributions. They also show the most dynamic response to rapid concentration changes with a t_10-90 time constant of around 1.5 s compared to the other sensors (&gt; 4 s).</p> Graphical Abstract <p></p>