Gas-phase atmospheric organic sulfur mixing ratios from expedition ARTofMELT, Arctic Ocean, 2023
Megan Willis, Cort Zang
This dataset consists of gas phase organic sulfur (dimethyl sulfide, methanethiol, dimethyl sulfoxide, and dimethyl sulfone) mixing ratios measured using a chemical ionization time-of-flight mass spectrometer during the Atmospheric Rivers and the onset of sea ice MELT (ARTofMELT) expedition.
This data was measured as part of the Secondary Marine Aerosol precursors and Links to aerosol growth at ice-melT onset in the Arctic (SMALTA, NSF Award 2211153) project during ARTofMELT.
The ARTofMELT expedition took place onboard Swedish icebreaker Oden, in the Fram Strait region between Greenland and Svalbard during May and June of 2023.
AtmosphereGas phaseArcticArctic OceanDimethyl sulfideDMSMethanethiolDimethyl sulfoxideDMSODimethyl sulfoneFram StraitARTofMELTARTofMELT 2023 expeditionIcebreaker Oden
Visit web siteName
Version
Citation
References
Zang CL, Willis MD (2025) Deployment and evaluation of an NH₄⁺/H₃O⁺ reagent ion switching chemical ionization mass spectrometer for the detection of reduced and oxygenated gas-phase organic compounds. Atmospheric Measurement Techniques 18:17–35. https://doi.org/10.5194/amt-18-17-2025
Data description
The dataset is providded as an ict octet stream (application file).
Comments
Methods
Step 1 (Ethical Research Practices)
This dataset is composed of atmospheric composition observations made using a chemical ionization mass spectrometer sampling ambient air from the 4th deck of I/B Oden during the ARTofMELT expedition in the Fram Strait. The Swedish Polar Research Secretariat (SPRS) managed logistics, permitting and permissions for the ARTofMELT expedition. SPRS obtained "Marine Research" under the United Nations Convention on the Law of the Sea (UNCLOS) rules, from Norway and Denmark. Data from ARTofMELT is being published openly in accordance with SPRS open data policy.
Step 2
Ambient air was sampled to a Vocus chemical ionization time-of-flight mass spectrometer (Vocus-CI-ToFMS) from ~20m above sea level through a heated (30°C), ~50ft long, 3/8in outer diameter PFA tube at 15 liters per minute. The Vocus-CI-ToFMS alternated between hydronium (H3O+) ionization and ammonium (NH4+) ionization modes at 15-minute intervals as described in Zang and Willis, 2025. To prevent interactions of sampled compounds with stainless steel surfaces, we bored through the stainless steel connection on the Vocus-CI-ToFMS inlet with 0.25 inch (6.35 mm) outer diameter (O.D.), 0.15 inch (3.8 mm) inner diameter (I.D.) perfluoroalkoxy (PFA) tubing to the polyether ether ketone inlet capillary. The Vocus-CI-ToFMS was mounted to the floor of the sampling container on I/B Oden's 4th deck using five high deflection vibrational isolation feet (Barry Controls Model 2K2-BA-90) and connected on one side to an open wall using two wire rope isolators (Enidine Model WR6-850-10-E). Vocus-CI-ToFMS raw time resolution was adjusted from 5 Hz to 2 Hz sampling frequency on May 12th, 2023.
Raw mass spectra from the Vocus-CI-ToFMS were collected using Acquility (version 2.3.18) and TofDaq (version 1.99) (Tofwerk AG). Mass spectra were mass calibrated, peak fit, and time integrated using Tofware (version 3.3.0) (Tofwerk AG and Aerodyne Research Inc.). Time-integrated high-resolution ion signals were exported for further analysis in Python (version 3.13.2).
Step 3
Background periods when the Vocus-CI-ToFMS inlet was overflowed with zero air are used to background subtract ion signals detected in this dataset. Background periods are determined using a 5-minute rolling average of the zero air pollution system state (i.e., once the pollution system was active for 5-minutes, this is classified as a background period). We smooth the background signal using a 60 second rolling average and linearly interpolate over non-background periods. The smoothed and interpolated background data is subtracted from the raw dataset to obtain a background subtracted dataset. From the raw (unsmoothed) background data, we calculated a detection limit as 3 times the 2-minute rolling standard deviation of the background. This is linearly interpolated over periods without background data to obtain a time dependent detection limit based on the most recent background period. After background subtraction, we clean the background subtracted ambient dataset for a variety of problems including reagent ion switching hysteresis described in Zang and Willis 2025, and unstable ion source periods (diagnosed using the standard deviation of the regulating voltage).
Step 4
We calculated sensitivities of the Vocus-CI-ToFMS using both H3O+ and NH4+ to a variety of compounds including dimethyl sulfide (DMS), dimethyl sulfoxide (DMSO) and methanethiol (MeSH) using multipoint calibrations from standardized gas cylinders (Apel Reimer Environmental Inc.). During the measurement period, we used a suite of deuterated internal standards to monitor instrument sensitivity changes over time. Our internal standards consisted of acetone(d6), dimethyl sulfide(d3), 2-hexanone(d4), and 1,3,5-trimethylbenzene(d12). Internal standards were injected at the top of our inlet at a rate of 5 sccm into the main inlet flow of 15 LPM.
Project
NSF Award 2211153 (Secondary Marine Aerosol precursors and Links to aerosol growth at ice-melT onset in the Arctic (SMALTA))
Publisher
NSF Arctic Data Center
License
Creative Commons Attribution 4.0 International License (CC BY 4.0)
First name
Megan
Last name or organisation
Willis
Email address
Address
Department of Chemistry; Colorado State University
Postal code
80524
City
Fort Collins
Province or state
Colorado
Country
United States of America
Dataset language
English
Begin date
2023-05-10
End date
2023-06-12