Waging The War Against DNAPLsby
Original Publication Date: January 1996
Until quite recently, if groundwater on a site proved to be contaminated by DNAPLs (dense, non-aqueous phase liquids, such as chlorinated solvents, pesticides and PCBs), it was generally regarded as the environmental equivalent of a terminal cancer diagnosis. DNAPLs, like their histological counterpart, are very difficult to detect, much less treat, and will become a continual source of toxic contamination for an aquifer, which can be spread by the act of sampling them. A growing awareness that continued remediation of DNAPLs using conventional methods such as pump-and-treat will not remediate the DNAPL but only cause it to release contamination at a higher rate has led researchers and industry to seek innovative techniques for the detection, containment and treatment of this hidden contamination. Recent advances in these efforts deserve more attention and further investigation and provide reasons for optimism.
Although awareness of DNAPLs dates back over two decades, the pervasiveness of the problem has only recently become evident. The EPA, for example, reported in 1993 that in over 60% of sites where organic contamination has occurred the likely source is DNAPLs and that 70% of all Superfund sites with groundwater contamination have NAPLS (non-aqueous phase liquids of which DNAPLs are a sub-category). A majority of these non-Superfund sites with a potential DNAPL problem are manufacturing facilities, ranging from large multi-million dollar operations to small tool and die shops.
The problem for manufacturers both large and small is that federal and state regulations require groundwater to be cleaned to demanding standards and in some cases equal to drinking water. Yet groundwater contaminants such as DNAPLs can form insoluble and highly mobile pools that defy all conventional cleanup methods. Despite this fact, failure to adequately address DNAPL contamination can subject the business owner to regulatory penalties and liabilities that can result from the spread of the contamination to adjacent properties.
DNAPLs (dense, non-aqueous phase liquids) also have been called "sinkers" and "toxic blobs", colorful descriptions which capture two of the most prevalent features of these compounds. DNAPLs are "sinkers" because they are heavier than water and sink until they hit an aquitard, a change in soil type or density. As a result, DNAPL movements are directed more by gravity than the flow of groundwater. If the base of the aquifer slopes in one direction, then the DNAPL will flow in the same direction seeking the lowest point. The direction of DNAPL movement may be counter to the groundwater flow, and therefore counter to the focus of traditional groundwater investigation programs.
Once they reach the low point of their descent, typically an aquitard, DNAPLs will form pools or "toxic blobs" of pure waste that slowly dissolve in the surrounding groundwater in the form of small contamination plumes. The DNAPL pool may even release contaminants in a groundwater plume that is independent of other groundwater plumes at the site. Since DNAPLs have very low solubility points, they can continue releasing small quantities of contaminants into the groundwater for centuries.
DNAPLs can include compounds such as: chlorinated solvents used to clean and de-grease machinery, creosol based wood treating oils, coal tar wastes, pesticides, PCBs, chloroform, carbon tet, methylene chloride and trichloride ethylene, to mention only a few of the more prevalent ones. Many of these chemicals are commonly used in a wide variety of manufacturing businesses. What makes DNAPLs so dangerous is the fact that they all degrade to other compounds which are even more insidious.
The definition of a DNAPL, however, depends on a combination of factors, which includes the compound, the concentration of the compound in the soils or groundwater, the original rate of release of the compound and the make-up of the saturated soils. Because of the interplay of these various factors, the determination of whether an instance of contamination comes under the technical definition of DNAPL requires the expertise of a knowledgeable consultant.
DNAPLs are a form of NAPLs (non-aqueous phase liquids). This group of groundwater contaminants also includes LNAPLs (light non-aqueous phase liquids, comprised of petroleum products such as gasoline and diesel oil) which float on top of the groundwater and make them easier to characterize, map and remediate. In addition to DNAPLs and LNAPLs, there also are NNAPLs ( neutrally buoyant non-aqueous phase liquids) which can be present at varying depths of the groundwater. As with DNAPLs, the exact characterization of each form of NAPL is determined by a combination of factors that include the compounds, release rates and site characteristics. Although all forms of NAPLs share low solubility points to varying degrees, only DNAPLs are heavier than water, which makes them very difficult to find and to effectively remediate.
Typically the first indication that a site may have a potential DNAPL problem occurs during a Phase I site assessment, which is routinely required prior to the sale of industrial property or financing additional construction. A key element of a Phase I assessment is a detailed review of the site's history and use, including a record of all chemicals that may have been released on the site. If the Phase I assessment finds that a significant quantity of DNAPL compounds (de-greasers, solvents, PCBs, etc.) were dumped on the site, then further analysis will be required.
Even after a determination is made that DNAPL compounds were released, the likelihood of a DNAPL groundwater problem will still depend on such factors as the total quantity of the release, the period of time over which the release occurred, and the make-up of the saturated soils. For example, if a small shop owner dumped 7 or 8 five-gallon containers of solvent over a ten year period, there is very little likelihood of a DNAPL problem. On the other hand, if a large manufacturing operation had routinely released hundreds of barrels of cleaning solvents into the ground each year, then a DNAPL problem is likely, although still far from certain, and further tests are required.
If the site history reveals significant releases of DNAPL compounds, then the next phase of site characterization should include actual soil and groundwater tests to determine if DNAPLs are present in the soil and whether they are limited to just the soil. Any soils or groundwater investigation must address the issue by sampling in places most likely to contain a DNAPL pool (due to their dense properties, these locations may differ from those that would be examined during a standard soils or groundwater investigation). Caution is the key to successfully conducting these tests, since some procedures such as the drilling of test wells can aggravate a DNAPL problem if not properly done, by causing the DNAPL to become mobile and spreading to other parts of the aquifer.
For this reason, it is best to begin with non-invasive tests such as a soil gas survey which can determine the presence of DNAPL solvents in the unsaturated zone before it has actually entered the groundwater. This test can be supplemented by such other site characterization techniques as surface geo-physical surveys and air photo interpretation of stressed vegetation. The combination of these tests can help determine the likelihood of a DNAPL problem as well as providing a general idea of the problem's primary location and likely dispersion.
Once a determination has been made that a DNAPL problem appears likely, then more invasive procedures are required to measure the severity and extent of the contamination. The most common techniques include borings and the drilling of monitoring wells at various strategic points and at different depths on the site. To reduce the risk of accidentally aggravating the problem, it is advisable to do this drilling moving from the outside perimeters of the site in toward the known areas of contamination or hot spots.
Computer modeling of the site and the "behavioral" characteristics of the DNAPL compounds can assist in estimating the location and migration of the pool and resulting plume without additional invasive testing, thus reducing the likelihood of releasing DNAPLs during investigation. Modeling also helps to further focus sampling efforts to intersect the most likely pathways of contamination and perhaps even locate the pools of DNAPL. To effectively model the site requires an understanding of the density, viscosity and hydraulic conductivity of the aquifer, the angle or slope of the aquitard and their combined effect on the velocity of the DNAPL. But even with this information, locating the actual sources of DNAPL contamination can take weeks or months of drilling test wells as well as conducting multiple model runs.
Although no single remediation technique has yet proven itself effective in the treatment of all DNAPL contamination situations, considerable success has been achieved in the containment of the problem. Today, if a site is shown to have a DNAPL problem, the first thing to do is determine the extent of the problem and then implement measures to contain and reduce the size of the contamination plumes. Often, portions of an aquifer can be restored to clean-up standards, even if the entire aquifer can not be cleaned up. These containment efforts usually include a pump and treat system or in some cases tile and drain systems (Love Canal, Niagara Falls, NY) which have been successful in containment of DNAPL contamination.
By containing the contamination plume, the area dependent on engineered systems and institutional controls will be reduced, which will decrease long-term operating costs and limit any subsequent liability resulting from dispersion of the DNAPLs into the groundwater of adjoining properties.
Progress also has been made in efforts to actually clean up DNAPL contamination. Most of these new cleanup procedures have either combined or modified existing remediation technologies. Some of the more promising DNAPL remediation techniques include:
The so-called "triple train approach" advocated by Mark Mercer of the Hazardous Site Control Division of the EPA, according to which the pure pools of DNAPL are first pumped and treated, then secondary recovery techniques are used to clean the residual wastes that adhere to soil and rock, and finally the cleanup is completed with pump-and-treat and/or bioremediation of the ground water.
The "washout", a variation of the foregoing technique, has been developed at SUNY (Buffalo) and uses surfactants (soaps) to wash out DNAPLs from aquifers. With this technique contaminated water is pumped to the surface where surfactants are added to it, and then pumped back into the aquifer. Over time, the surfactants flush the DNAPLs out of the aquifer and air stripping is used to remove the contaminants.
Despite the success of several initial tests, neither of these techniques, nor any other, has yet been shown to work on all sites. Current clean-up techniques continue to be site specific and largely experimental. So the best advice for treating a potential DNAPL problem today is to react quickly, determine the nature and scope of the contamination, review the applicability of existing DNAPL remediation techniques and if no current techniques are appropriate, then contain the problem and wait until remediation technology catches up.
Waiting will also enable the current regulatory environment to catch up to the latest scientific findings regarding DNAPLs and their remediation. Regulatory agencies still assume that nearly every site is contaminated with unremediateable DNAPLs (hence the value of knowledgeable consultants who can define DNAPLs properly). To make matters worse, some of these agencies will not allow the use of any remediation technology that requires the introduction of materials into the aquifer (viewing it as contamination) no matter how well meaning or potentially effective. Consequently, it may be best to wait for the regulatory agencies to come to the realization that DNAPLs are not ubiquitous and that they can be remediated.
Copyright 1996 by Thomas Wackerman. All rights reserved.