Fast Liquid Chromatography Coupled with Tandem Mass Spectrometry for the Analysis of Vanillic and Syringic Acids in Ice Cores

The development of new analytical systems and the improvement of the existing ones to obtain high-resolution measurements of chemical markers in samples from ice cores, is one of the main challenges the paleoclimatic scientific community is facing. Different chemical species can be used as markers for tracking emission sources or specific environmental processes. Although some markers, such as methane sulfonic acid (a proxy of marine productivity), are commonly used, there is a lack of data on other organic tracers in ice cores, making their continuous analysis analytically challenging. Here, we present an innovative combination of fast liquid chromatography coupled with tandem mass spectrometry (FLC-MS/MS) to continuously determine organic markers in ice cores. After specific optimization, this approach was applied to the quantification of vanillic and syringic acids, two specific markers for biomass burning. Using the validated method, detection limits of 3.6 and 4.6 pg mL–1 for vanillic and syringic acids, respectively, were achieved. Thanks to the coupling of FLC-MS/MS with the continuous flow analytical system, we obtained one measurement every 30 s, which corresponds to a sampling resolution of a sample every 1.5 cm with a melting rate of 3.0 cm min–1. To check the robustness of the method, we analyzed two parallel sticks of an alpine ice core over more than 5 h. Vanillic acid was found with concentrations in the range of picograms per milliliter, suggesting the combustion of coniferous trees, which are found throughout the Italian Alps.


Details of Continuous Flow Analysis systems
. Schematic of the continuous flow analysis (CFA) system optimized for coupling with the fast liquid chromatography-mass spectrometer (FLC-MS/MS). Table S1. Description of the peristaltic pumps and tubing used in the CFA system. Table S2. Optimized mass spectrometric parameters. DP: Declustering Potential, EP: Entrance Potential, CE: Collision Energy, CXP: Cell Exit Potential Table S3. Chromatographic parameters used to evaluate the instrumental performance using a standard solution of vanillic (VA) and syringic (SyA) acids at a concentration of 100 pg mL -1 and the internal standard labelled vanillin (VAH*) at a concentration of 1 ng mL-1. Average values of peak area (in counts), peak width (in seconds), peak asymmetry and MS identification points for the first (C1) column (n=5) and the second (C2) column (n=5) and total average values together with the standard deviation in brackets are reported.

Details of the Continuous Flow Analysis system
The melting system is comprised of an aluminum anodized square-based melting head (external dimension of 32x32 mm 2 and internal dimension of 21x21 mm 2 ), specifically chosen after previous tests that were conducted with different melt head designs. The characteristic high walls separate the outer from the inner sections reducing the possibility of meltwater crossover, and any contamination potentially coming from the outermost melt water flow. Furthermore, the anodized aluminum guarantees chemical stability as well as having excellent thermal properties at a reasonable cost. The melting unit is thermostated between 30° and 33°C by an electrical heater coupled to a temperature PT100 sensor, located in an upright freezer kept at −20 °C. High-purity perfluoroalkoxy (PFA) tubing (1/16 in. OD, 0.02 in. ID) and standard 1/4-28 low-pressure fittings (both from Upchurch, USA) were used throughout the system. Ice was melted at a constant rate of 3 cm min -1 adjusting the temperature of the melt-head with a PID controller. The ice core samples have a square cross section as we receive a cut sub sample of the ice core. The stick is loaded into acrylic sample guides that are loaded onto the melter head that keep the sample upright. A weight is added to the top of the core to apply pressure to ensure smooth melting of the sample.
The melting line passes through a 10-port low-pressure switching valve (Cheminert, Vici) to allow rapid switching 1) to the sample line during ice core analysis, or 2) to ultra-pure water, during the wash phase. The switching valve was also used to load calibration standards before and after ice core analysis. Table S1 describes the overall CFA pumping system (peristaltic pumps and tubing). A first peristaltic pump (Ismatec ISM942) is set at 3.4 mL min -1 to 1) draw the sample from the melt-head, and 2) to remove external contaminated meltwater coming from the external surfaces of the ice core.
After an enclosed debubbler (c. 200 μL internal volume), for removing air bubbles from the liquid sample stream, the flow is split into two lines. The first line with a flow rate of 1.8 mL min -1 was collected by a fraction collector to obtain discrete samples. The second line with a flow rate of 1.6 mL min -1 brought the sample stream into a manifold. Here, the flow is split in three lines that are redirected to the same second peristaltic pump. In the first line, the insoluble particle counts (Abakus Klotz, Germany) and electrolytic conductivity (AmberScience, USA) are continuously determined online at a flow rate of 1.3 mL min -1 (green line in the Fig. S1). The second line (red in the Fig. S1) is directed to the injection system of the FLC-MS/MS to ensure a constant flow of 0.26 mL min -1 . A third line (blue stopper in the Fig. S1) is available for other analyses but is not used in this study. The use of fluoroelastomer (FPM) tubing in the FLC-MS/MS line is necessary because serious contamination problems were found using other peristaltic tubing materials (i.e. Tygon, Solvaflex, Marprene) when determining vanillic and syringic acids.
To guarantee a completely bubble-free stream to the chromatographic system, the sample passed through another debubbler (c. 100 μL internal volume) before being redirected again to the second peristaltic pump. This is because the ice can get stuck in the sample holder during the melting resulting in the introduction of air into the sample stream. There is switching valve, that allows for a quick change between sample to ultrapure water (UPW) to prevent large air bubbles being introduced, but despite these precautions bubbles can still enter the system. Therefore, the benefits offered by the second debubbler were considered higher than any potential contamination risk because the FLC system needs to be protected from air bubbles. The debubblers were placed under a small laminar flow hood to reduce potential contamination from the lab atmosphere. From our preliminary tests, no significant contamination was observed. Figure S1. Schematic of the continuous flow analysis (CFA) system optimized for coupling with the fast liquid chromatography-mass spectrometer (FLC-MS/MS). Table S1. Description of peristaltic pumps and tubing using in the CFA system.   (2)