| 24 Apr 2026
Insufficient mass spectrometric detection of synthesized peroxy acids from α-pinene ozonolysis
Abstract. Biogenic volatile organic compounds (BVOCs) are major precursors of secondary organic aerosol (SOA) and new particle formation (NPF), and therefore play an important role in the climate system, altering the abundance of cloud condensation nuclei (CCN). Ozonolysis of the most atmospherically abundant monoterpene, α-pinene, generates RO2 radicals which undergo autoxidation, resulting in the formation of oxygenated organic molecules (OOMs) with low volatility, an essential step in nucleation and early particle growth. However, quantitative interpretation of widely used mass spectrometric OOM measurements remains limited by reagent-ion selectivity and the lack of authentic monomeric standards, an issue that is particularly important for hydroperoxides and peroxy acids, which constitute a significant fraction of autoxidation products. Here, we synthesize two α-pinene-derived monomeric OOM standards, peroxy norpinonic acid (PNPA; C9H14O4) and peroxy pinonic acid (PPA; C10H16O4), and confirm their structures by 1H and 13C NMR. We then evaluate their detectability using a MION-Orbitrap operated with nitrate (NO3−) and uronium (CH5N2O+) chemical ionization and compare these gas-phase schemes to heated electrospray ionization (H-ESI). NMR shows that freshly prepared standards are dominated by peroxy acids and contain only a fraction of the corresponding carboxylic acids, whereas Orbitrap measurements consistently yield substantially lower peroxy-to-carboxylic-acid ratios. These ratios vary strongly across ionization modes, with greater apparent peroxy acid loss under harder (de)protonation and improved, though still incomplete, preservation under softer nitrate and uronium adduct formation, indicating that the decomposition originates during ionization. Notably, uronium provides significantly higher sensitivity for these moderately oxygenated compounds, complementing nitrate’s strong selectivity toward highly oxygenated molecules (HOMs). Together, our results suggest that peroxy acids formed via α-pinene autoxidation may be systematically under-quantified by commonly used mass spectrometric approaches, with implications for O:C assignments, volatility-basis-set derivations, and inferred atmospheric process rates of nucleation and early particle growth.
This manuscript addresses an important issue in atmospheric measurements by examining the limitations of widely used mass spectrometric OOM measurements, particularly those arising from reagent-ion selectivity and the lack of authentic monomeric standards for hydroperoxides and peroxy acids. The topic is timely and relevant, given the central role of OOMs in secondary organic aerosol formation and new particle growth. The results suggest that peroxy acids may be systematically underestimated due to ionization-related effects, which has important implications for interpreting OOM measurements. However, several aspects of interpretation and the broader atmospheric implications require further clarification.
The central conclusion of the manuscript is that peroxy acids formed during α-pinene oxidation are substantially underdetected by nitrate-CIMS due to decomposition or fragmentation associated with the ionization process. However, the evidence presented does not fully exclude alternative explanations related to the chemical stability of the synthesized standards during storage, handling, or sampling into the instrument. Can the authors provide additional experimental evidence demonstrating that the observed discrepancy between the NMR-derived concentrations and the mass spectrometric response originates specifically from the ionization step? For example, are there measurements under different ionization conditions or inlet residence times that would help isolate the source of the signal loss? Thermal decomposition, wall losses, or inlet-related processes could contribute to the observed discrepancies. Given the known instability of peroxy compounds, can the temperature of ion transfer tube affects the stability of the cluster? Also, I suppose the total signal of the products in nitrate CIMS is the sum of normalized signal intensities of deprotonated as well as the cluster. Does addition of all the related signals (deprotonated + cluster) reduces the discrepancy observed between NMR and MS-derived ratios?
The study relies on two monomeric compounds - peroxy norpinonic acid (PNPA) and peroxy pinonic acid (PPA). This point is particularly important because the broader atmospheric implications discussed in the manuscript rely on the assumption that the behavior of the investigated standards is representative of peroxy acids present in α-pinene ozonolysis and other HOM-forming systems. This study relies on a limited number of synthesized standards, and it is unclear to what extent the results can be generalized to the broader class of highly oxygenated atmospheric peroxy acids formed during α-pinene ozonolysis.
Minor comments
1. Expansion of term HOM should be consistently defined upon first use. For example, in Line 17, HOM is expanded as ‘highly oxygenated molecules’.
2. Figure 5 middle panel figure, why the signal(m/z 247 and 231) persist in the background of the uronium ionization experiments?