Iron speciation in soil size fractions under different land uses

Highlights

• We investigated Fe speciation in fine sand (FSa) and fine silt plus clay (FSi + Cl).

• The nature of Fe-SOM interactions vary with soil particle size and land use.

• FSa fractions were characterized by more crystalline Fe phases, including goethite.

• FSi + Cl fractions exhibit the presence of ferrihydrite.

• SOM in the FSi + Cl fraction is more thermodynamically stable than in FSa.

Abstract

Iron (Fe) (oxyhydr)oxides represent a significant phase for the organic carbon (OC) stabilization. Due to their high surface areas, short-range-ordered Fe minerals, like ferrihydrite, show a higher ability to stabilize OC than crystalline secondary minerals, like lepidocrocite, goethite, and magnetite. However, how Fe phases and their crystallinity relate to soil organic matter (SOM) composition is still not completely known. We investigated Fe solid phase speciation in two soil particle size fractions (i.e., fine sand — FSa — and fine silt and clay — FSi + Cl —), isolated from coniferous (CF) and broadleaved forest soils (BF), grassland soils (GL), and technosols (TS). All samples were characterized by Fe K-edge extended X-ray absorption fine structure (EXAFS) and X-ray diffraction, and a subset by ⁵⁷Fe Mössbauer spectroscopy. Further, ramped combustion thermal analyses (coupled differential scanning calorimetry (DSC) and CO₂ evolved gas analysis (CO₂-EGA)) were used to evaluated SOM stability. With the only exception of TS, goethite was the main Fe phase in FSa fractions, whereas less crystalline phases (i.e., ferrihydrite, based on EXAFS) dominated the FSi + Cl fractions. The proportion of goethite and ferrihydrite in both fractions decreased with increasing OC content, while that of Fe(III)-SOM in FSa increased with increasing OC content. Mössbauer and EXAFS data clearly indicated a presence of hematite in GL soils. Our data suggest that more crystalline Fe forms, like goethite and hematite, may be important for OC abundance in the FSa fraction and in soils with high OC contents, like GL. Thermal analysis showed the dominance of mineral associated organic matter in low-OC soils, and of plant residues in high-OC soils. As a whole, we posit that the FSi + Cl fractions contain Fe phases of less crystallinity because of presumed association with SOM, and that SOM in the FSi + Cl fraction is also more thermodynamically stable than in FSa, although differences are observed across land uses. Our observations suggest that the nature of Fe-SOM interactions can vary substantially with soil particle size and land use, which has important implication for SOM persistence.

Introduction

Differences in soil organic carbon (SOC) stocks across ecosystems are mainly linked to climate and edaphic factors (Gray and Bishop, 2016, Jobbagy and Jackson, 2000). SOC stocks are also highly influenced by changes in land cover and use (Guo and Gifford, 2002, Kooch et al., 2012, Lal, 2010, Post and Kwon, 2000, Smith et al., 2012, Stockmann et al., 2015). Land use and management have a significant impact on both soil organic matter (SOM) molecular composition and its interaction with mineral surfaces (Rumpel et al., 2009, Shen et al., 2018, Wang et al., 2016). Forest and grassland soils usually have the largest organic carbon (OC) stocks among land covers (Carter et al., 1998, Conant et al., 2001, Guo and Gifford, 2002, McNally et al., 2017, Post and Kwon, 2000, Soussana et al., 2004).

Iron (Fe) (oxyhydr)oxides are widespread in soils across climates and ecosystems, though their abundances vary widely. Because land use changes can dramatically alter soil physical, chemical and biological properties, they can also influence Fe distribution and biogeochemistry and, consequently, the potential of soil to accumulate OC. In fact, a wide number of studies have indicated that Fe oxide phases strongly influence SOC storage (Torn et al., 1997, Percival et al., 2000, Eusterhues et al., 2005, Rasmussen et al., 2005).

The way in which soil properties contribute to the mechanisms of SOC stabilization (e.g., Wagai and Mayer, 2007), such as aggregation or organo-mineral association (von Lützow et al., 2008), has been extensively studied, although a common ground for their formation or composition has not been established (Prechtel et al., 2009). The proportion of different mineral classes in soils, including crystalline clays, low crystallinity short-range-order (SRO) oxides and organo-metal complexes, has been often estimated using selective mineral dissolution methods, although they are generally applied in a contrasting way. Advanced spectroscopic techniques represent a faster and non-destructive alternative to sequential extractions for the quantification of Fe mineral species in soil particle size fractions (Heckman et al., 2018).

SOM stabilization mechanisms have been mainly investigated through soil fractionation by particle size. In fact, SOM in the fine-silt and clay (FSi + Cl) fraction may be stabilized by weak-force interactions (e.g., Van der Waals or hydrogen bonds), association with Fe and aluminium (Al) oxides, or precipitation by polyvalent cations (von Lützow et al., 2006), whereas the fine sand (FSa) fraction, having a low specific surface area (SSA) and reactivity, seems to play a secondary role for OC stability in soils (Six et al., 1998). However, the role of this fraction in SOM accumulation may have been underestimated, at least in agricultural soils (Giannetta et al., 2020a), as its adsorption and cation exchange capacity can be enhanced following coatings with Fe oxides (Karlsson et al., 2008). The understanding of SOM stabilization mechanisms based on size fractionation cannot overlook an exhaustive characterization of both SOM and the minerals, including Fe phases. Beyond the strong relationships between OC stocks and Fe content, less is known about how both Fe phase abundance and species relate to SOM chemical composition. For example, redox fluctuations can drive both the formation and disruption of SRO Fe complexes, suggesting that SRO Fe phases are not necessarily expected to stabilize SOM over extended periods. Alternatively, crystalline Fe phases, that are less prone to reductive dissolution (Ginn et al., 2017), could promote long-term SOM stabilization (Hall et al., 2018), although the OC stocks they are able to protect is smaller due to their lower SSA.

Fe Extended X-ray Absorption Fine Structure (EXAFS) spectroscopy provides both physical and chemical information at the molecular level which allow a better understanding of interactions between SOC and Fe (Giannetta et al., 2020a, Giannetta et al., 2020b, Giannetta et al., 2022), although its application to bulk soils and soil fractions still represents a data analysis challenge. Mössbauer spectroscopy has been used to constrain the EXAFS fitting on selected samples and to provide more accurate information about the crystallinity of Fe phases. In addition, thermal analysis for the characterization of SOM in soil size fractions can integrate the Fe speciation, unravelling the mechanisms of SOM stabilization through interactions with minerals (i.e., the energy input needed to break the bonding of SOM with mineral surfaces and the corresponding energy released or consumed).

Here, coupling particle size fractionation with EXAFS analyses, Mössbauer spectroscopy and X-ray diffraction (XRD), we assessed the relationship between Fe speciation and thermal indices of SOM stability in FSa and FSi + Cl fractions isolated from soils under different land uses, including coniferous forest soils (CF), broadleaved forest soils (BF), grassland soils (GL), and technosols (TS). The stability of SOM in both fractions was assessed by thermogravimetry (TG) and differential scanning calorimetry (DSC) coupled with carbon dioxide (CO₂) evolved gas analysis (CO₂–EGA). Both TG and DSC are generally used to study the stability of SOM (Lopez-Capel et al., 2005, Plante et al., 2009), determining, for example, the energy as heat necessary for SOM thermal oxidation (i.e., the energy barrier to SOM decomposition). The protection by organo-mineral interaction vs. chemical recalcitrance across land uses can drive the variability found in both Fe speciation and thermal stability of SOM (Williams et al., 2018). We hypothesize that (1) the finer fraction (FSi + Cl) would contain Fe phases of less crystallinity because of presumed association with SOM; and (2) SOM in the FSi + Cl fraction is more thermodynamically stable than in FSa, albeit depending on land uses.