Kurtz, A.C.; Derry, L.A.; Chadwick, O.A. 2002. Germanium-silicon fractionation in the weathering environment. Geochimica et Cosmochimica Acta 66, 9.
Problem statement. Trace elements can be a useful aid in understanding soil weathering processes. There is a need to better understand Germanium behavior and Ge/Si fractionation. How are the two related? Also, previous studies have identified a low Ge/Si ratio in streams (Mortlock & Froelich 1987; Murnane and Stallard 1990). It was known that Si exports to streams, but it was not known where the Ge was going.
Goals. Attempt to understand Ge behavior in the soil weathering environment and Ge/Si fractionation. Where is the extra Ge going?
Study area and background. A sequence of lava flows in Hawaii, ranging from 0.3 ka to 4100 ka. What happens to the Ge/Si as the lavas and soils eventually weather? Ge is a pseudoisotope of Si (Azam and Volcani 1981), meaning it can substitute readily for Si. Ge is a trace element, at ~1 ppm in rocks. As the rocks weather, Ge/Si fractionation occurs. Fractionation can also occur as soils weather. Si is reapidly depleted by weathering, and Al is easily leached (Fig. 2). By 20 kyr, primary minerals have weathered away, replaced by noncrystallines such as allophone. These slowly recrystalize over >1 Ma, forming secondary kaolin (a fine white clay formed from the weathering of aluminous materials) and crystalline sesquioxides (3 oxygens, such as alumina). Kaolin and sesquioxides dominate the older sites (<2 µm fraction). Argillite (clay stone) is present below. Saprolites (soft, disintegrated rock) can also be present.
Methods. Measured Ge with ICP-MS (Mortlock & Froelich 1996). A lot of good detail in the methods section.
Results. According to table 3, the Ge/Si ratio increases with weathering. Fig. 3: scatterplot of SiO2 vs. Ge. This is not showing the ratio; rather it shows that Ge abundance is positively related to SiO2 concentration. Conversely, there seems to be an inverse relation between Ge and Fe2O2.
All the older soils tend to have Ge/Si near 10, but the young soils are around 3. The young soils tend to be Si enriched, while the older soils tend to be Ge enriched (relative to Si, but Ge concentration is much lower than in young soils).
The authors next set out to figure out why Ge seems to concentrate in older soils. They had three plausible explanations: precipitation of secondary aluminosilicate clays, Fe oxides, and the accumulation of organic matter. Three chemical extractions from the soils were performed. First was AOD, which presumably extracted Ge from noncrystalline aluminosilicates and Fe-Al sesquioxides. Second, DC extracted Ge from crystalline Fe- and Al-sesquioxides. Third, NaOH extracted Ge from kaolin and biogenic opal. Ge extractions from these steps would presumably tell the researchers what the proportion of Ge was for each step. I won’t go into too much detail on the numbers. What they did find was that Ge concentration seems to increase with weathering, up to a point. Eventually as deep weathering continues, Ge will decrease as well. Thus, it seems that Ge is enriched for a while, but then progressively drops off. Figure 4 shows this: the Ge/Si increases for a while, but eventually drops off. It was found that organic accumulation has nothing to do with Ge enrichment. Fe seems to have little to do with it either. Rather, it was found that secondary soil silicate fractions tend to have high Ge/Si fractions.
Discussion. I think this is a novel method to determine the amount of weathering in a soil. For my research, this method could prove useful: since it is very difficult to remove clays from small particles of biogenic Si (BSi), it could be possible to measure the Ge/Si for each sample. That way I could have a reliable estimate of the amount of clay Si input into the sample. Assuming of course, that clays and BSi have differing Ge/Si.
This technique could be coupled with Al/Si. Methods similar to those used in radiogenic isotope geochronology could be used. Geochronologists will often compare the ratio of a pair of isotopes against another. For example, it is known that 238U has a shorter radioactive half-life than 235U. With time, one would expect the 235U/238U to increase. By itself, this can be a good chronometer. But as they say, two is always better than one. Geochronologists can add another chronometer: 232Th/230Th. Just as with 235U/238U, 232Th/230Th will increase with time. Thus, the two chronometers offer a robust check against one another. In my research, the Ge/Si could be plotted against the Al/Si. Both are potential proxies for soil weathering, but together they can be more reliable.
Conclusion. This study was performed in humid tropical soils. I wonder how the Ge/Si would behave in a temperate semi-arid environment.