Tejada, Rayan2023-02-162023-02-162022-12https://hdl.handle.net/11299/252493University of Minnesota M.S. thesis. December 2022. Major: Chemical Engineering. Advisor: Steven Sternberg. 1 computer file (PDF); vi, 52 pages.Low-cost particulate matter sensors (LCPMSs) could provide significant insight into air quality data with their ability to be placed virtually anywhere, short sampling time, and cost to build. However, LCPMSs are also known to significantly overestimate particle counts when the relative humidity (RH) is above 65%. It is widely considered that the hygroscopic growth of aerosols is the cause. Hygroscopicity of PM can be described by a single parameter, symbolized as K, and was used in a previous study (Di Antonio et al., 2018) to correct LCPMS data with promising results. However, the study assumed ambient PM to be a pure substance, however, it is often found to be a complex mixture of organic and inorganic chemical species. This study tested if a statistically derived empirical value of K, referred to as “ambient K”, could improve representing the RH influence on LCPMSs. Ambient K is defined as the statistically best-fitting value for several experimental observations of hygroscopy and makes no assumptions on the number of species in ambient PM. Ambient K was graphically demonstrated to be more representative of the experimentally observed RH error compared to assuming K, while having the same statistical performance as conventionally assuming K. Varying observations of hygroscopic behavior among multiple sensors provided strong evidence of multiple chemical species in the observed ambient PM.enAerosolsAir QualityHygroscopicityLow Cost SensorsParticulate MatterPM2.5Thesis or Dissertation