American Chemical Society
ac7b02004_si_001.pdf (123.68 kB)

126 264 Assigned Chemical Formulas from an Atmospheric Pressure Photoionization 9.4 T Fourier Transform Positive Ion Cyclotron Resonance Mass Spectrum

Download (123.68 kB)
Version 2 2017-10-12, 19:43
Version 1 2017-10-11, 18:17
journal contribution
posted on 2017-10-11, 00:00 authored by Logan C. Krajewski, Ryan P. Rodgers, Alan G. Marshall
Here, we present atmospheric pressure photoionization (APPI) Fourier transform ion cyclotron resonance (FTICR) mass analysis of a volcanic asphalt sample by acquiring data for 20 Da wide mass segments across a 1000 Da range, stitched into a single composite mass spectrum, and compare to a broad-band mass spectrum for the same sample. The segmented spectrum contained 170 000 peaks with magnitude greater than 6σ of the root-mean-square (rms) baseline noise, for which 126 264 unique elemental compositions could be assigned. Approximately two-thirds of those compositions represent monoisotopic (i.e., chemically different) species. That complexity is higher than that for any previously reported mass spectrum and almost 3 times greater than that obtained from the corresponding broad-band spectrum (59 015). For the segmented mass spectrum, the signal-to-noise ratio (S/N) was significantly higher throughout the spectrum, but especially at the lower and upper ends of mass distribution relative to that of the near-Gaussian broad-band mass distribution. Despite this S/N improvement, mass measurement accuracy was noticeably improved only at lower masses. The increased S/N did, however, yield a higher number of peaks and higher dynamic range throughout the entire segmented spectrum relative to the conventional broad-band spectrum. The additional assigned peaks include higher heteroatom species, as well as additional radicals and isotopologues. Segmenting can require a significant investment in data acquisition and analysis time over broad-band spectroscopy (∼1775% in this case) making it best suited for targeted analysis and/or when complete compositional coverage is important. Finally, the present segmented spectrum contains, to our knowledge, more assigned peaks than any spectrum of any kind (e.g., UV–vis, infrared, microwave, magnetic resonance, etc.).