[Chemical diversity of dissolved organic matter revealed by ultra performance liquid chromatography-Fourier transform ion cyclotron resonance mass spectrometry].

Dissolved organic matter (DOM) is a highly complex and heterogeneous mixture that exists in various environments, including rivers, oceans, soils, and atmospheric aerosols. DOM plays a crucial role in biogeochemical cycles and significantly influences the environment by regulating water quality, changing the climate, and transporting pollutants. Therefore, clarifying the detailed molecular composition of DOM is essential to obtain a better understanding of its physical and chemical properties, thereby enabling further elucidation of its biogeochemical behavior. In this study, Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) combined with quadrupole detection (QPD) was used to conduct the online ultra performance liquid chromatography (UPLC)-MS analysis of DOM in water, aerosol, and soil samples collected in Tianjin, China. The samples were extracted with pure water and filtered through a glass fiber membrane (0.45 μm). The DOM in the samples was then enriched by solid-phase extraction (SPE) and redissolved in water-acetonitrile (1∶1, v/v) at mass concentration of 200 mg/L for the LC-MS experiments. The mobile phases used for UPLC were water containing 0.1% (v/v) formic acid (A) and acetonitrile containing 0.1% (v/v) formic acid (B). The gradient elution procedure was as follows: 0-5 min, 0B; 5-11 min, 0B-95%B; 11-25 min, 95%B; 25-28 min, 95%B-0B; 28-30 min, 0B. The flow rate was 0.1 mL/min, and the injection volume was 10 μL. The UV wavelength was set at 274 nm. MS detection was performed in negative electrospray ionization (ESI(-)) mode with a capillary voltage of 5.0 kV, and the MS data were collected in broadband ( 150-1000) and QPD modes. The transient data size was set to 2M, the free induction decay signal length was 0.74 s, and the ion accumulation time was 0.030 s. Four chromatographic peaks were observed in the chromatograms. The first peak was identified as salt adduct compounds containing sodium formate. The three other peaks contained complex components, such as oxygen-rich, unsaturated tannin-like compounds, as well as low-oxygen, highly saturated lignin-like and protein/amino-like compounds. UPLC-FT-ICR MS was suitable for assigning the detailed elemental compositions of the DOM samples. UPLC effectively improved the ionization efficiency of difficult-to-ionize compounds and enhanced the detection accuracy of MS. Indeed, MS peaks with a mass difference of as small as 3.4 mDa were well identified. A total of 12027, 15593, and 8029 peaks in the mass spectra of the water, aerosol, and soil samples, respectively, were assigned to known elemental formulae. Peaks Ⅱ and Ⅲ were hydrophilic components mainly including CHNO and CHO compounds. Compared with peak Ⅱ, peak Ⅲ exhibited a significant increase in CHNOS and CHOS, indicating that UPLC exerted a certain separation effect on these compounds. Furthermore, the aerosol samples contained a higher concentration of sulfur-containing compounds than the water and soil samples, primarily because of the abundance of organic sulfates present in atmospheric and cloud water. Data processing and graphic visualization revealed that the unique components in the water samples mainly appeared in the area of 0.1<O/C<0.5 and 1.0 <H/C<1.7. The compounds detected were low-oxygen and highly condensed lignin-like compounds. The unique components in the aerosol samples appeared in the area of 0.4<O/C<1.0 and 1.5<H/C<2.0, and were classified as carbohydrates. The unique components in the hydrophilic fraction of the soil samples were found in the area of 0.6<O/C<1.0 and 0.5<H/C<1.0, and were determined to be tannin-like compounds. By contrast, the components in the hydrophobic fraction were similar to those found in the water samples and appeared in the region containing lignin-like compounds. In summary, this study proposed a novel analytical protocol to characterize DOM from different ecosystems using UPLC-FT-ICR MS. This method could separate DOM components using UPLC with eluents of different polarities and analyze them using high-resolution FT-ICR MS to reveal their molecular compositions and possible chemical types. This protocol offers solid technical support for the comprehensive profiling of DOM at the molecular level.