Comprehensive assessment of dissolved organic matter processing in the Amazon River and its major tributaries revealed by positive and negative electrospray mass spectrometry and NMR spectroscopy.

Rivers are natural biogeochemical systems shaping the fates of dissolved organic matter (DOM) from leaving soils to reaching the oceans. This study focuses on Amazon basin DOM processing employing negative and positive electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry (ESI[±] FT-ICR MS) and nuclear magnetic resonance spectroscopy (NMR) to reveal effects of major processes on the compositional space and structural characteristics of black, white and clear water systems. These include non-conservative mixing at the confluences of (1) Solimões and the Negro River, (2) the Amazon River and the Madeira River, and (3) in-stream processing of Amazon River DOM between the Madeira River and the Tapajós River. The Negro River (black water) supplies more highly oxygenated and high molecular weight compounds, whereas the Solimões and Madeira Rivers (white water) contribute more CHNO and CHOS molecules to the Amazon River main stem. Aliphatic CHO and abundant CHNO compounds prevail in Tapajos River DOM (clear water), likely originating from primary production. Sorption onto particles and heterotrophic microbial degradation are probably the principal mechanisms for the observed changes in DOM composition in the Amazon River and its tributaries.

Automated microextraction sample preparation coupled on-line to FT-ICR-MS: application to desalting and concentration of river and marine dissolved organic matter.

Sample preparation procedures are in most cases sample- and time-consuming and commonly require the use of a large amount of solvents. Automation in this regard can optimize the minimal-needed injection volume and the solvent consumption will be efficiently reduced. A new fully automated sample desalting and pre-concentration technique employing microextraction by packed sorbents (MEPS) cartridges is implemented and coupled to an ion cyclotron resonance Fourier-transform mass spectrometer (ICR-FT/MS). The performance of non-target mass spectrometric analysis is compared for the automated versus off-line sample preparation for several samples of aqueous natural organic matter. This approach can be generalized for any metabolite profiling or metabolome analysis of biological materials but was optimized herein using a well characterized but highly complex organic mixture: a surface water and its well-characterized natural organic matter and a marine sample having a highly salt charge and enabling to validate the presented automatic system for salty samples. The analysis of Suwannee River water showed selective C18-MEPS enrichment of chemical signatures with average H/C and O/C elemental ratios and loss of both highly polar and highly aromatic structures from the original sample. Automated on-line application to marine samples showed desalting and different chemical signatures from surface to bottom water. Relative comparison of structural footprints with the C18-concentration/desalting procedure however enabled to demonstrate that the surface water film was more concentrated in surface-active components of natural (fatty acids) and anthropogenic origin (sulfur-containing surfactants). Overall, the relative standard deviation distribution in terms of peak intensity was improved by automating the proposed on-line method.