Some years ago while in graduate school at the University of New Hampshire, I was presented the opportunity by my advisor, Dr. Chris Evans, to solve a hard soil mystery resting mostly unnoticed except to those who had to work in it. Found on steeper slopes under coniferous forest, for builders, the Success series can be troublesome. It’s effort enough to dig any soil, but the Success soil is especially hard to dig. It contains an indurated horizon. It’s a sandy soil with a fair amount of gravel and cobbles and it’s derived from glacial till. Although solving the cement mystery was sort of fun, the bigger goal was to learn more about weathering processes and products associated with the genesis of those New Hampshire soils. The effort was eventually published in a peer-reviewed journal article (Freeland and Evans, 1993).

Understanding soil weathering processes and products requires a general knowledge of the local geologic framework and some low-temperature aqueous geochemistry. In my case, the stakes seemed small, but in some instances lack of knowledge about soil weathering can turn deadly as we’ve seen with mass poisoning by arsenic groundwater contamination in Bangladesh. There, for decades, shallow tube wells were installed by international development agencies in an effort to provide clean drinking water. In the Ganges River delta, the fresh alluvium eroded from the tectonically active mountains to the north weather rapidly in the warm humid climate.

Note to those responsible for geology department curriculum: teach about weathering and include it in the groundwater program.

Here is USGS geochemist Andrea Foster delivering an excellent lecture on arsenic in groundwater (YouTube).

The figure to the right shows a thin-section micrograph of the Success soil with some stressed quartz grains and some finer textured “gunk” between the grains. It was the “gunk” coating t the grains, technically called cutans, we were most interested in.

Silica and Quartz Cement

There were competing theories about the source of induration in the Success soil. One idea was silica cement or “silcrete.” Silcrete is more common arid climates where it forms “durapans,” also known as “duricrusts”. New Hampshire’s 40-inches or so (100+ cm) of annual precipitation, made this option unlikely, but we tested for it.

Another possible “cementing mechanism” was pressure solution of quartz grains. This process, as I understand it, essentially involves molecules in the solid phase obtaining higher kinetic energy under pressure and at differential stress conditions at grain contacts, setting up a diffusion gradient. The molecules can then migrate enough to deform grain boundaries and cause grains to “knit together” at points of contact. It’s a subject that brings structural geologists and geochemists together. Some more discussion of the pressure solution phenomenon, which seems not well understood or even clearly defined, is available here and here. If pressure solution of quartz grains (silica) was the cementing agent, then we should have been able to dissolve that kind of “glue” using sodium hydroxide solution.

Clay
Another theory had it that clay made the coarser grains stick together. In that case, because clay absorbs water, we should have been able to soften up the clay with a “slake test,” which is just soaking a soil clod in water to see if it falls apart. Soil scientist Ray Archuletta demonstrates a slake test in this YouTube video. The “biological glues” he refers to are mainly fats, waxes and resins derived from plant roots and other soil organisms.

Sesquioxides
Gibbsite, ferrihydrite, hematite, and goethite are common iron and aluminum oxides found as weathering products in acidic soils (Sposito, 1989). They are collectively known as “sesquioxides,” which is the general term soil scientists use for oxides of aluminum and iron. These weathering products have been identified as cementing agents in Scottish podzols (Romans, 1962), Ohio Fragiudalfs (Hallmark and Smeck, 1979). Sesquioxides have also been identified as common cementing agents in Spodosols (Birkeland, 1984).

Organic Matter including Fats Waxes and Resins
The cemented horizons (Bsm, BCm) of the Success soil are below the top (A) horizon and appear to be mostly light colored and low in organic matter content. Still, we felt we needed to rule out organic glues as the cementing agent. In Ray Archuletta’s slake test video linked above, Ray attributes the bonding of the water-stable clod to “hydrophobic organic glues.”

Organo-metallic Complexes
Soil humus is made up of long-lasting organic materials that have undergone limited microbial processing to form polymers (Sposito, 1989). Two categories of humic substances are humic and fulvic acids. They are both variable in chemical compositional and are distinguished on the basis of differential solubility. The significant aspect of both of them for my study is their role as chelators. They form mobile complexes with metals and may have been part of the process moving metals to the cemented horizons of the Success soils. We could test for this kind of cement by chemically “digesting” the organic matter.

Allophane and Imogolite
Lastly, we wanted to investigate the possibility of clay-like minerals first identified in volcanic soils (Andisols) of Japan and New Zealand. Allophane and imogolite are chemically similar but differ in degree of atomic order, with imogolite the more organized of the two. They are both aluminosilicates with an Al:Si ratio of 2:1. Allophane is spheroidal and imogolite is rolled up into a tube.

Solvents and Reagents

These were the chemicals used to try to disaggregate the indurated soils, along with the materials each reagent would attack:

Silica or Aluminum compounds – 0.05 M NaOH
Clay (fragipan) – water
Organic matter – 10% boiling hydrogen peroxide solution matter
Iron or aluminum compounds – 0.1 M HCl
Allophane and imogolite – 0.2 M ammonium oxalate at pH 3 because allophane and imogolite are the only significant sources of oxalate extractable Si (Parfitt and Hemni, 1982; Farmer et al. 1983).

The greatest extent of disintegration came from treatment with ammonium oxalate and to a lesser extent, destruction of organic matter with boiling hydrogen peroxide. The hydrogen peroxide digestion produces a lot of effervescence, which probably helped break other kinds of adhesive bonds between grains. Based on that, we had reason to believe that allophane and or imogolite (allophanic materials) were the primary cementing agent and to a lesser extent, organic matter.

We would go on to additional tests (more about them later) to rule out or confirm the presence of allophanic materials. The chemical tests went a long way to identify the weathering products in the Success soil. From there, we’d have to work “backwards” through plausible weathering processes that produced those secondary minerals and the soil morphologies seen in the field and in micrographs.

The overall process for developing the kind of soil we were working with, soils produced from quartz rich, sandy parent materials in cool, humid, acidic. coniferous forested conditions is called podsolization. The process has been a topic of debate for many years with some scientists quite convinced of their favorite theories.

I came away with a strong opinion about how the Success soils “worked” in terms of weathering processes and products, but continue to suspect there are often multiple sets of processes and sequences that lead to similar end results.

References

Anderson, H.A., M.L. Berrow, V.C. Farmer, J.D. Russell, and A.D. Walker. 1982. A reassessment of Podzol formation processes. Journal of Soil Science 33:125-136.

Birkeland, P.W. 1984. Soils and Geomorphology. Oxford University Press, Inc. New York.

Farmer, V.C., J.D. Russell, and M.L. Berrow. 1983. Extraction of inorganic forms of translocated Al, Fe, and Si from a podzol B horizon. Jurnal of Soil Science 34:571-576.

Freeland, J.A., and C.V. Evans. 1993. Soil Science Society of America Journal. 57:183-191

Hallmark, C.T., and N.E. Smeck. 1979. The effect of extractable aluminum, iron, and silicon on strength and bonding of fragipans of northeastern Ohio. Soil Science Society of America Journal. 43:145-150.

Parfitt, R.L., and T. Hemni. 1982. Comparison of an oxalate-extraction method and an infrared spectroscopic method for determining allophane in soil clays. Soil Science and Plant Nutrition 28:183-190.

Romans, J.C.C. 1962. The origin of the indurated Bx horizon of podzolic soils in north-east Scotland. Journal of Soil Science 13:141-147.

Sposito, G. 1989. The Chemistry of Soils. Oxford University Press, New York.