Global distribution of nonexchangeable stable hydrogen isotope ratios of topsoil clay fractions

Equilibrium isotope fractionation factors of H exchange between steam and soil clay fractions

“… Notes : The δ 2 H n values for α ex‐w = 1 and the CEC pot were taken from previous studies 23,24 . “Untreated” and “treated” indicate samples not subjected and subjected to clay separation, including removal of Fe oxides and carbonates, reduction in SOM, and dispersion of the remaining material. …”

“…This is corroborated by the increasing χ e values with increasing latitude in this group of soil clay fractions (Table 1), because it is known that the χ e values decrease in the order: swellable three-layer clay minerals > nonswellable three-layer clay minerals (illite) > two-layer clay minerals > mica. 23,24 The α ex-w values of the group of soil clay fractions with a CEC pot > 400 mmol c kg À1 , presumably dominated by swellable three-layer clay minerals, correlated with the δ 2 H p values, but the slope was flatter than that for the group of soil clay fractions with a CEC pot of 160-400 mmol c kg À1 (Figure 3). We attribute this finding to the fact that the group of soil clay fractions with a CEC pot > 400 mmol c kg À1 contained a less heterogeneous mixture of minerals, possibly dominated by a single three-layer clay mineral.…”

“…We used the H isotope ratio measurement, clay separation, and steam equilibration methods of Merseburger et al 23,24 Briefly, after weighing the cooled capsules, we measured the δ 2 H values using an elemental analyzer-pyrolysis-isotope ratio mass spectrometer (Flash 2000 HTCÀDelta V Advantage, Thermo Fisher, Waltham, MA, USA) in a chromium-filled Al 2 O 3 reactor at 1250 C. We corrected small memory effects with the help of a pool-wise memory correction algorithm using one or two pools. 75 For normalization, we used…”

“…A study of an Argentinian climosequence demonstrated a weaker correlation of the δ 2 H values of local precipitation with the δ 2 H n values of bulk soil than with the δ 2 H n values of demineralized SOM 10,27 . Merseburger et al 24 showed that one reason for the weaker correlation of the δ 2 H values of local precipitation with the δ 2 H n values of bulk soil than of SOM is the weaker correlation of the δ 2 H values of local precipitation with the δ 2 H n values of clay fractions than of SOM assuming the same constant α ex‐w values for all samples. The larger scatter in the relationship between the δ 2 H n values of soil clay fractions and the δ 2 H values of local precipitation was attributed to different isotope fractionation between ambient water‐H and H of clay minerals (α min‐water ) during their formation 28 and a minor contribution of rock‐derived clay minerals that were not formed under the current climatic conditions.…”

“…67,68 All samples were dried and stored in the dark for 1-28 years. Further properties of the used soil samples, including X-ray powder diffractometric patterns of six samples, can be found in Merseburger et al 24 Additional X-ray powder diffractometric patterns have been published for sample PA-A9 60 and from soils nearby the locations from where our three Ecuadorian samples were collected. 57…”