土壤有机质（SOM）的生物地球化学循环不仅与铁元素紧密相关，并且两者的相互作用是影响SOM稳定性的重要因素。然而，不仅铁矿物间的化学转化错综复杂，且SOM是由许多结构和成分未知的分子通过复杂的相互作用组合而成，因此铁矿物对SOM的化学保护还没有在分子层面阐释清楚。近年来，电喷射离子化傅里叶变换离子回旋共振质谱（ESI-FT-ICR MS）技术的飞速发展，为我们在分子层面更精准分析SOM提供了可能。本研究采用ESI-FT-ICR MS仪器，以四种典型铁矿物（水铁矿、赤铁矿、硫化亚铁矿和黄铁矿）和天然可溶性有机质（DOM）为研究对象，从分子层面探究由于铁矿物对DOM吸附所引发的分子分馏作用。研究的主要结论如下： （1）当以单位质量吸附剂（μg C g-1）为计量标准时，平衡吸附量按照以下顺序排列：水铁矿>硫化亚铁矿>赤铁矿>黄铁矿；但当以单位比表面积吸附剂（μg C m-2）来计量时，平衡吸附量顺序变为：硫化亚铁矿>黄铁矿>赤铁矿>水铁矿。水铁矿排序的巨大变化，证明了水铁矿的高比表面积（309.8 m2 g-1）是主导DOM吸附的一个重要因素；同时前者的排序与四种矿物吸附前Zeta电位的排序一致，且Zeta电位与logqe （μg C g-1）显著相关（R2>0.96，P<0.001），这都表明静电作用在吸附过程中也起到了一定的作用； （2）水铁矿和硫化亚铁矿比赤铁矿和黄铁矿对DOM有更强的吸附能力和分子分馏作用，这种选择性吸附在低浓度(Ce < 20 mg L-1)时较明显，但随着平衡浓度的升高逐渐减弱并逐渐达到稳定，这可能是由于随着浓度的升高铁矿物对有机质的选择性吸附逐渐达饱和； （3）水铁矿和硫化亚铁矿对DOM的分子分馏体现在以下两个方面。在分子量方面：水铁矿、硫化亚铁矿均选择性吸附分子量较大（>500 Da）的分子，这种对高分子物质呈现出选择性吸附的主要原因可能是疏水作用；在分子化学组成方面：不饱和度（DBE）高、标化碳氧化值（NOSC）高的物质或富氧物质（主要是含氧芳香碳，多酚以及含羧基化合物）更倾向于吸附到水铁矿和硫化亚铁矿上，而DBE低或含氧官能团少的物质（主要是含氧量少的脂肪碳）会被留在水相。铁矿物对芳香碳和多酚的高选择性是由于苯环与铁间的电子转移以及酚类有机物与铁间的配位络合，对羧基的高选择性是因为羧基与铁间的配体交换、螯合、离子架桥和氢键等作用，这些作用可以促使DOM与铁形成更稳定的络合物。因而，铁矿物对DBE高、NOSC高或O原子数多的物质呈现出高选择性的最根本的原因在于：铁矿物能够与这些物质中的芳香碳、酚类以及羧基发生相互作用。这些发现能给土壤有机质中带有各种不同含氧官能团的芳香碳和脂肪碳的稳定性研究提供更有说服力的证据。

The biogeochemical cycles of iron and soil organic matter (SOM) are strongly associated, and the interaction between them is an important factor affecting the stability of SOM especially at the redox interfaces. However, the role of iron minerals with various redox mineralogies in chemical protection of SOM has not been clearly characterized at the molecular level because of the complication of SOM and the redox change of iron. Recently, the rapid development of electrospray ionization coupled with Fourier transform ion cyclotron resonance mass spectrometry (ESI-FT-ICR MS) has provided a more accurate analysis of SOM at the molecular level. Successfully, it was used in this study to investigate the molecular fractionation of DOM by adsorption onto four representative redox iron minerals (ferrihydrite, hematite, iron sulfide, and pyrite). The main conclusions of the study are as follows: (1) The sorption affinity (normalized to per gram adsorbent, μg C g-1) to DOM was in the order of ferrihydrite > iron sulfide > hematite > pyrite. However, the sorption affinity (normalized to per square specific surface area, μg C m-2) followed the order of iron sulfide > pyrite > hematite > ferrihydrite. This demonstrated the importance of the high specific surface area (309.8 m2 g-1) of ferrihydrite in the DOM adsorption. And the former sequence was the same as the Zeta potential of iron minerals before adsorption which was highly correlated with logqe (μg C g-1) (R2>0.96, P<0.001), which both suggested the influence of electrostatic interaction in DOM adsorption. (2) Ferrihydrite and iron sulfide exhibited more pronounced adsorption affinity and molecular fractionation than hematite and pyrite. This selective adsorption was more pronounced at low concentrations (Ce < 20 mg L-1) but decreased with increasing equilibrium concentrations and gradually stabilized, which might be due to the finite adsorption capacity of iron minerals. (3) The molecular fractionation of DOM induced by adsorption by iron minerals was reflected both in molecular weight and molecular component. In the aspect of molecular weight fractionation, ferrihydrite, hematite and iron sulfide exhibited stronger fractionation to high molecular weight compounds, which might due to the hydrophobic interaction. As for molecular component fractionation, compounds high in unsaturation (DBE) and nominal oxidation state of carbon (NOSC) or rich in oxygen (mainly polycyclic aromatic carbon, polyphenols and carboxyl groups) were preferentially bound to ferrihydrite and iron sulfide, leaving compounds low in unsaturation or poor in oxygenated groups (mainly aliphatic carbon with few oxygen atoms) in solution. The high selectivity of iron minerals to aromatic carbons might be ascribed to the coordination complexation with phenolic carbon as well as electron donor-receptor interactions with aromatic rings. The high selectivity to carboxyl group is due to the ligand exchange, chelation, ion bridging, and hydrogen bonding with iron which generates more stable complexes successfully. And the fundamental reason of high selectivity for molecular with high DBE, high NOSC, or more O atoms was the high selectivity to aromatic carbons, phenols, and carboxyl groups. These findings would provide new insight into the stability of aromatic and aliphatic carbon with various oxygenated groups coupled to redox iron transformation in soils along different depth or with redox fluctuation.