Distribution of Contaminated Soil: trimodal soil distributions
By the earthDr!
How do soils become contaminated?

There are many different types of chemicals that can contaminate soil. Some chemicals are gases at room temperature. Some chemicals are liquids at room temperature such as the metal mercury. Other chemicals are solids at room temperature such as the metal lead. Some chemicals, in liquid form, can completely dissolve in water such as ethanol - drinking alcohol. Other chemicals, in liquid form, can dissolve only very slightly such as benzene. Some chemicals, in solid form, dissolve in water but to a limited extent. Other chemicals in solid form dissolve fairly easily, but with limitation such as table salt. Some gases can dissolve in water as carbon dioxide does to a limited extent.

Before we can understand how soil becomes contaminated by these different types of chemicals, soils need to be defined. A soil can be divided up into different fractions. The soil contains mineral particles (the mineral fraction), water (soil solution or groundwater), air (soil atmosphere), and organic matter (containing many different humic acids and fulvic acids). There are many different types of minerals. Minerals are just the unique way a particular group of chemical elements are arranged in a rock or soil particle. Most of the elements in the mineral fraction of the soil are derived from the weathering of rock. The mineral fraction of a soil is not necessarily just minute particles of the larger, local parent rocks. As larger rocks weather into smaller rocks, both physical and chemical weathering occur. Larger rocks are physically ground down into smaller rocks. Water, wind or other geologic processes can grind rock into smaller rock. Water can be in either a liquid or solid state. A turbulent stream can grind and polish larger rocks into smaller ones much like a rock tumbler. A glacier can grind and polish larger rocks into smaller ones. The mineralogy does not change by this physical weathering as a larger rock is ground down into a smaller particle. Physical weathering can have an effect on chemical weathering, though. Chemical weathering can change the mineralogy of these minerals since they are now more intimately exposed to chemical weathering by having been ground down into smaller particle sizes by the action of physical weathering.

The surface area per unit mass increases as a particle is ground to a smaller size (this phenomenon will be discussed later). This increasing surface area allows for greater chemical weathering of the smaller-sized particles relative to larger-sized particles. As the surface area per unit mass increases, chemical weathering can have a more pronounced effect on the mineralogy of the smaller sized particles since more surface area is exposed (as a result of the increased surface area per unit mass as a particle size decreases) to the actions of chemical weathering.

Rocks can dissolve into water. A metal such as aluminum, within a mineral, may dissolve into the water and then, another metal, say, iron which may be present in the soil solution or ground water may actually replace (the atom of iron can go from the solution into the mineral replacing) the aluminum in the exact position formerly occupied by the aluminum atom in the prior mineral assemblage. This substitution/replacement of iron for aluminum will change the surface charge of the mineral and therefore, change the surface chemistry of the mineral.

As rocks dissolve into water, the concentration of dissolved salts in the water can increase. At a later date, these salts can precipitate out of the water forming new types of minerals (secondary minerals) as water is lost to evaporation (such as in the upper ranges of the capillary fringe). Over time these secondary minerals may subsequently dissolve and later form new minerals within the soil. Dissolution and precipitation, dissolution and precipitation, and so on are an ongoing process.

The mineral fraction of the soil can become contaminated. The soil water (soil solution) can become contaminated. The soil atmosphere can become contaminated. It is reasonable to assume that when the mineral fraction of the soil becomes contaminated, then the soil solution, the soil organic matter, and the soil atmosphere also becomes contaminated since there is such an intimate intermingling of the soil mineral fraction with the soil solution, the soil organic matter, and the soil atmosphere.

The first example chemical which we will investigate is a liquid at room temperature and is only slightly soluble in water. There are a number of ways that this chemical can contaminate the soil. This chemical can directly contaminate soil as a separate phase, in liquid form (produsol); it can dissolve in water and then, the contaminated water can contaminate the soil (solusol); and, a vapor cloud of this contaminant can contaminate the soil (vaposol). However, the upper limit concentration for those soils contaminated by contaminated water (solusol) when compared with soils contaminated directly by product (produsol) is substantially less contaminated. When soils become contaminated through contact with contaminated water (solusol), it is possible to calculate an upper concentration limit of the contamination in the solusols. Don't think that these calculations are exact, but they do achieve a result close enough, to get a good approximation, to real world situations. These calculations require certain assumptions that achieve an approximate result.

The figure on this page illustrates the distribution of contaminated and clean soils at a site that could be contaminated by such slightly-soluble products as TCE, benzene, toluene, and/or the xylene isomers. This class of contaminants is only slightly soluble in water which will affect the distribution of contamination in the soil (overlapping versus non-overlapping modes). This figure depicts a trimodal distribution. Each mode represents a different population. The mode at the right of the figure represents those soils contaminated by contact with product. These soils I call produsols. The x-axis of this figure represents the concentration of contamination in the soil. The y-axis of this figure represents the frequency of soils detected at a concentration of contamination that can range from not contaminated to highly contaminated. This mode, at the right of the figure, represents the greatest concentration of contamination.

The second mode from the right contains those soils contaminated by polluted water or by contaminant vapors. Those soils contaminated by dissolved product (contaminated water) I call solusols; those soils contaminated by vapors I call vaposols. Please note that the concentrations of contamination in solusols and vaposols are two to four orders of magnitude less contaminated than are produsols. The only difference between vaposols and solusols is the method that the soil becomes contaminated (genesis). Both vaposols and solusols are morphologically indistinguishable. The last mode, depicted to the left is labeled clean soil. Some of those soils labeled clean are not really clean because the collected sample contains a mix of clean soil and some level of soil contamination.

This figure is only a conceptual representation of what will be found in the real world. The logic to support the belief that a trimodal distribution of clean and contaminated soils exists is based upon the calculation of the upper limit of soil contamination for solusols and vaposols; and, the empirical definition of the lower concentration limit along with the calculation of the upper concentration limit for produsols.

The upper limit of for dissolved product contamination of solusols is derived from the following assumptions: all soils contain water; contaminants from both the liquid phase and vapor may be able to dissolve in this water; and, contaminants may be able to sorb (adsorb or absorb) to the soil particle from the soil solution or ground water. Based upon these assumptions contaminant contribution in the soil solution/ground water, sorption to the soil particulates, and the soil atmosphere can be calculated.

Obviously, product directly contaminating soil should result in the greatest concentrations of soil contamination. There have been numerous experiments to identify the amount of product that can be retained in variously textured soils. Rough estimates of product retention can also be had by using the porosity of the soil to calculate the volume of product retained in the soil. It should be noted that the porosity of a clay is greater than is the porosity of a sand. Don't falsely interpret the smaller-diameter of the soil pores in a clay to mean that the clay is less porous than a sand and; therefore, should retain less product. The only limitation to the degree of soil contamination would be the amount of pore space available for product entry and trapping. However, don't falsely conclude, that a soil of higher porosity (clay) might be able to retain more product than a soil of lower native porosity (sand). But remember, I said "the amount of available pore space" and this can be influenced by the amount of soil moisture above the water table.

Identifying that a trimodal distribution exists is not just an academic endeavor. This distribution has a real import. Solusols and vaposols compared to produsols contain substantially less contaminant mass. As such, it is not difficult to exhaust these lesser concentrations of soil contamination. The flushing of contaminants from soils can be accomplished by either clean air or clean water. Flushing of contaminants from produsols can require many hundreds of flushes with clean water to achieve an acceptable level of soil contamination. Flushing with clean air, provided the contaminant has a high enough vapor pressure, may be accomplished in fewer flushes. However, flushing contaminants from solusols or vaposols, either with clean air or clean water, can be accomplished with many fewer flushes since the contaminant mass is 1/100 to 1/10,000 that of produsols.

Remember that this figure is only a conceptual representation of that found in the real world. Whenever, a soil sample is collected remote from a contaminated site, it is most often void of contamination. The detection of contaminated soils are generally some combination of contaminated soils with clean soils mixed in with it. As soil sampling proceeds closer to the release or spill, a greater mass of contamination will be present in the soil sample. It is not only possible to bias toward one mode or the other, but it is warranted. However, all to often bias is introduced without a conscious decision being made to achieve it.

It should be understood that these three modes represent three different soil populations: clean soil; slightly contaminated soil (solusols/vaposols); and, highly contaminated soil (produsols). What value is there in summing the contaminant concentration in each of the soils, without regard to the population that a particular soil concentration represents, and then, dividing this sum by the number of soil samples to determine an average soil contaminant concentration? Isn't there a greater value in first assigning contaminant concentration ranges, albeit rough contaminant concentration ranges, to the solusols/vaposols and produsols and then, determining the spatial extent of each range and the beginning of the clean soils?