Humic acids from particulate organic matter in the Saguenay Fjord and the St. Lawrence Estuary investigated by advanced solid-state NMR

摘要

Detailed structural information on two humic acids extracted from two sinking particulate matter samples at a water depth of 20 m in the Saguenay Fjord (F-20-HA) and the St. Lawrence Estuary (E-20-HA) (Canada), was obtained by advanced solid-state NMR. Spectral-editing analyses provided numerous structural details rarely reported in geochemical studies. The NMR data account almost quantitatively for the elemental compositions. The two humic acids were found to be quite similar, consisting of four main structural components: peptides (ca. 39 +/- 3% vs. 34 +/- 3% of all C for E-20-HA and F-20-HA, respectively); aliphatic chains, 14-20 carbons long (ca. 25 +/- 5% vs. 17 +/- 5% of all C); aromatic structures (ca. 17 +/- 2% vs. 26 +/- 2% of all C); and sugar rings (14 +/- 2% vs. 15 +/- 2% of all C). Peptides were identified by C-13{N-14} SPIDER NMR, which selects signals of carbons bonded to nitrogen, and by dipolar DEPT, which selects CH-group signals, in particular the NCH band of peptides. The SPIDER spectra also indicate that heterocycles constitute a significant fraction of the aromatic structures. The aliphatic (CH2)(n) chains, which are highly mobile, contain at least one double bond per two chains and end in methyl groups. H-1 spin diffusion NMR experiments showed that these mobile aliphatic chains are in close (< 10 nm) proximity to the other structural components. A major bacterial contribution to these two samples could explain why the samples, which have different dominant organic matter sources (terrestrial vs. marine), are similar to each other as well as to degraded algae and particles from other waters. The NMR data suggest structures containing mobile lipids in close proximity to peptides and carbohydrates (e.g., peptidoglycan) as found in bacterial cell walls. Measured yields of muramic acid and D-amino acids confirmed the presence of bacterial cell wall components in the studied samples. (c) 2007 Elsevier Ltd. All rights reserved.