The development of a soil profile, which may involve the dissolution of certain minerals and the leaching of ions from one location, and the subsequent precipitation of new minerals at another, reflects local climatic influences. An interesting example from a reactive transport perspective is the formation of so-called spodic soils, or spodosols.

Spodosols are a specific class of soils that are found primarily in cool, coniferous forests at high latitudes (e.g., boreal forests). Pine needles form a layer of soil litter on the forest floor. As rainwater infiltrates the litter, it leaches out various organic acids that subsequently migrate into the underlying soil column. These acids serve to leach Fe and Al from the soil by dissolving Fe(OH)₃ and Al(OH)₃ in the acidic water.

Moreover, the organic acids exhibit a tendency to form aqueous complexes with Fe and Al. The result is an increase the solubilities of Fe and Al minerals by tying up these cations in solution. However, as Fe- and Al-complexes are carried further down the soil column, the water chemistry is restored to less acidic conditions through reactions with other mineral phases (e.g., feldspars, clay minerals, carbonates). This reduces again the solubility of the hydroxides, leading to their subsequent precipitation. The overall process of spodosol formation thus results in the formation of a grayish, leached zone depleted in Fe and Al, underlain by a brown or red zone that is enriched in Fe and Al hydroxides.

PHREEQC can be used to study the spodosol formation problem and to examine hypotheses as to the specific mechanisms involved. Consider a simple case of a one-dimensional column of soil that is initially composed of K-feldspar and ferrihydrite, Fe(OH)₃. Now suppose water in equilibrium with atmospheric CO₂ begins entering the column. Furthermore, letís assume this water contains citric acid (HOOC-CH2COH-COOH-CH2COOH, or 2-Hydroxypropane-1,2,3-tricarboxylic acid) at a concentration of 10 mg/L as a surrogate for plant litter leachate. We have chosen citric acid because it is known to complex both Fe and Al in solution; the thermodynamic data for the formation of Fe-citrate complexes exist with the MINTEQ database supplied with the PHREEQC program (thus we neglect Al for this demonstration). Consider the following PHREEQC input file:

As an exercise, letís examine two scenarios, one with citric acid present at 10 mg/L in the influent and the other with no citric acid; all other attributes are held the same. Simulation results from running PHREEQC with these sets of input instructions are as follows

Figure 1. Distribution of mineral phases (mol/kg of water) along the soil column ("Solution" refers the cell number) with citric acid present in the influent after 1000 shifts. Note not only the leaching of Fe from the inlet end of the column, and its subsequent enrichment further downstream, but also the conversion of K-feldspar (microcline) into muscovite, kaolinite, and finally quartz

Figure 2. Distribution of mineral phases in the absence of citric acid after 1000 shifts; compare to Figure 1

Figure 3. Profiles of solid-phase Fe, as Fe(OH)₃, in the soil column after 1000 shifts, clearly illustrating the differences in the redistribution of Fe when citric acid is present versus absent. Leached zone is in Solutions 1-3; enriched zone in Solutions 4-7.

Figure 4. pH distribution in the soil column after 1000 shifts, both scenarios.

Question: Does the presence of citric acid redistribute the Fe as a result of complexation (Figure 3) or simply because of reduced pH? How can we decide?

Figure 5. Speciation of ferric iron in the presence of citric acid in the enriched zone after 1000 shifts. Note that approximately 97% of the dissolved iron is complexed with citrate. This greatly enhances the mobility of the iron in the soil column.

The degree to which ferrous iron is complexed by citrate, at least according to the MINTEQ thermodynamic data set, argues strongly that complexation probably plays more of a role in forming the simulated spodic soil than does the additional lowering of the pH offered by the presence of citric acid in the influent. This hypothesis could be tested by re-running the simulation with a non-complexing acid, such as HCl, at concentrations that reproduces the pH of the influent water with citric acid. This is left as an exercise for the interested student.

The example problem presented in this lecture assumed citric acid as a surrogate organic acid in the spodosol formation process. In reality, of course, a more complex mixture of multiple types of organic acids would likely play a role. Try to see if you would achieve similar results using other organic complexes in the MINTEQ database, such as butanoic or propanoic acid or EDTA. Also, what about redox effects? We have essentially ignored redox for this example. What if we assumed that the plant litter decomposes to a degree, creating an anaerobic environment in the soil solution where Fe²⁺ is the dominant form of iron, as opposed to Fe³⁺? How will this effect the simulation? Experiment with the input instructions and find out.

You are now ready to continue to

TOPIC D: SUMMARY AND FUTURE DIRECTIONS IN MODELING REACTIVE TRANSPORT.

         LECTURE 1: Conclusions and the Path Forward.

You may e-mail me questions and comments.

Walt W. McNab
E-mail address: mcnab1@llnl.gov.

Copyright Notice