Evaluating the performance of five different chemical ionization techniques for detecting gaseous oxygenated organic species
Matthieu Riva,Pekka Rantala,Jordan E. Krechmer,Otso Peräkylä,Yanjun Zhang,Liine Heikkinen,Olga Garmash,Chao Yan,Markku Kulmala,Douglas Worsnop,Mikael Ehn
DOI: https://doi.org/10.5194/amt-12-2403-2019
2018-01-01
Atmospheric Measurement Techniques Discussions
Abstract:Abstract. The impact of aerosols on climate and air quality remains poorly understooddue to multiple factors. One of the current limitations is the incompleteunderstanding of the contribution of oxygenated products, generated from thegas-phase oxidation of volatile organic compounds (VOCs), to aerosolformation. Indeed, atmospheric gaseous chemical processes yield thousands of(highly) oxygenated species, spanning a wide range of chemical formulas,functional groups and, consequently, volatilities. While recent massspectrometric developments have allowed extensive on-line detection of amyriad of oxygenated organic species, playing a central role in atmosphericchemistry, the detailed quantification and characterization of this diversegroup of compounds remains extremely challenging. To address this challenge,we evaluated the capability of current state-of-the-art mass spectrometersequipped with different chemical ionization sources to detect the oxidationproducts formed from α -Pinene ozonolysis under various conditions.Five different mass spectrometers were deployed simultaneously for a chamberstudy. Two chemical ionization atmospheric pressure interface time-of-flightmass spectrometers (CI-APi-TOF) with nitrate and amine reagent ionchemistries and an iodide chemical ionization time-of-flight massspectrometer (TOF-CIMS) were used. Additionally, a proton transfer reactiontime-of-flight mass spectrometer (PTR-TOF 8000) and a new “vocus” PTR-TOFwere also deployed. In the current study, we compared around 1000 differentcompounds between each of the five instruments, with the aim of determiningwhich oxygenated VOCs (OVOCs) the different methods were sensitive to andidentifying regions where two or more instruments were able to detectspecies with similar molecular formulae. We utilized a large variability inconditions (including different VOCs, ozone, NO x and OH scavengerconcentrations) in our newly constructed atmospheric simulation chamber fora comprehensive correlation analysis between all instruments. This analysis,combined with estimated concentrations for identified molecules in eachinstrument, yielded both expected and surprising results. As anticipatedbased on earlier studies, the PTR instruments were the only ones able tomeasure the precursor VOC, the iodide TOF-CIMS efficiently detected manysemi-volatile organic compounds (SVOCs) with three to five oxygen atoms, and thenitrate CI-APi-TOF was mainly sensitive to highly oxygenated organic (O u003e 5) molecules (HOMs). In addition, the vocus showed goodagreement with the iodide TOF-CIMS for the SVOC, including a range oforganonitrates. The amine CI-APi-TOF agreed well with the nitrate CI-APi-TOFfor HOM dimers. However, the loadings in our experiments caused the aminereagent ion to be considerably depleted, causing nonlinear responses formonomers. This study explores and highlights both benefits and limitationsof currently available chemical ionization mass spectrometry instrumentationfor characterizing the wide variety of OVOCs in the atmosphere. Whilespecifically shown for the case of α -Pinene ozonolysis, we expectour general findings to also be valid for a wide range of other VOC–oxidantsystems. As discussed in this study, no single instrument configuration canbe deemed better or worse than the others, as the optimal instrument for aparticular study ultimately depends on the specific target of the study.