Aerosols play a significant role in the atmosphere through affecting the radiative budget, cloud condensation nuclei activity, and visibility. They also cause adverse health effects leading to premature deaths. A major fraction of aerosols is organic matter (OM), which has a complex composition, is produced by various sources, and undergoes chemical transformation in the atmosphere, making its full characterization difficult. Biomass burning has become a major source of OM with an increasing effect due to more frequent large wildfires around the world resulting partly from climate change. In this thesis, Fourier transform infrared spectroscopy (FTIR) has been employed in six different projects as the main tool to characterize OM from various sources especially biomass burning in order to understand its formation and evolution in the atmosphere. This technique has been used to analyze the functional group (FG) composition of OM, specific spectral profiles from different sources, and biomass burning marker signatures in a nondestructive and cost-efficient manner. FTIR has also been combined with other analytical techniques. First, we used the spectral profiles in the aliphatic CH region of IR spectra to extract information about the molecular structure of atmospheric organic aerosols in terms of their mean carbon number and molecular weight. We found that urban, rural, and biomass burning aerosols have distinct mean molecular weights and carbon numbers. FTIR and aerosol mass spectrometry (AMS) were used in the second project to characterize burning aerosols in an environmental simulation chamber. We showed the agreement of these two instruments in terms of the OM mass concentration and elemental ratios (H:C, and O:C). We found that AMS spectra contained functional group information that agreed with that of FTIR even for moderately aged aerosols. In the third project, we used univariate and multivariate statistics to combine FTIR and AMS measurements to better understand and interpret the complex AMS mass spectra in terms of the FG composition and to estimate the high-time-resolution FG composition of combustion aerosols during the course of aging. In another environmental chamber study (the fourth project), we developed a procedure to estimate the often-neglected evolution of primary biomass burning aerosols during daytime and nighttime chemical agings. We found that at least 15 % of the primary aerosol mass undergoes chemical transformation at relatively short time scales in the order of a day in the atmosphere. Organic biomass burning markers were among the fastest decaying species, making the identification of aged biomass burning aerosols in the atmosphere challenging. In the fifth project, we used the biomass burning marker signatures in the FTIR spectra for the first time to identify and quantify atmospheric samples affected by wood smoke. The FTIR-based identification method agreed well with those using ion chromatography and satellite observations. This method, which is one of the few scalable to large air pollution monitoring networks, was applied to around 20,000 filters collected across the US in 2015 to estimate the impact of biomass burning. In the sixth project, we used FTIR to characterize OM emitted from different cookstoves and fuels. We found similarities between the spectra of unburned fuels and OM emissions and measured high abundances of aromatics and polycyclic aromatic hydrocarbons in the particulate emissions.