DNAPL History and Transport: A Summary
This post was prompted by interest in a lecture I attended as part of the Princeton Groundwater Pollution and Hydrogeology course (recommended!). North America’s most prominent DNAPL expert, John Cherry, spoke for approximately 5 hours about the topics I have summarized here. Cherry’s notes are copyrighted, as is the whole course, so I have paraphrased the take-home messages. An annotated list of references will follow in a subsequent post (soon)!
DNAPL is Dense Non-Aqueous Phase Liquid. The term refers to the group of groundwater contaminants that have a density greater than water (commonly cited as 1.0 g/cm3). These are a tricky group of compounds for contaminant hydrogeologists because the source tends to sink and can be difficult to find. DNAPLs include chlorinated solvents (tetrachloroethylene (PCE), trichloroethylene (TCE), trichloroethane (TCA), cis-1,2-dichloroethylene (DCE), vinyl chloride(VC)), creosote, coal tar, polychlorinated biphenyls (PCBs) and undiluted pesticides. The most widespread DNAPLs are chlorinated solvents because PCE is the main chemical used for dry cleaning, TCE is a heavily-used industrial solvent, and DCE and VC are the sequential daughter products. (VC degrades further into ethene and ethane, which are the desired end-products of in situ degradation of a chlorinated solvent.) Chlorinated solvents were reportedly developed in Germany in the late 19th Century; their use increased drastically during WWII. Until 1979, MSDSs for chlorinated solvents recommended disposal in dry soil because they were thought to volatilize.
Dissolved plumes caused by DNAPLs were discovered in the 1970s but DNAPL (the free phase, not dissolved phase) was not discovered until the mid-1980s. This is partially because monitoring wells are a poor method to detect DNAPL; it is rarely found in wells. Discovery was precipitated by legislation introduced during the previous decade: Safe Drinking Water Act (1974), Resource Conservation and Recovery Act (RCRA, 1976) and the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA, commonly known as Superfund, 1980). This legislation required sampling of municipal wells specifically for chlorinated solvents, which were summarily discovered in some drinking water systems. Unlike some other contaminants, such as my favorite, methyl tert-butyl ether (MTBE), chlorinated solvents have high taste and odor thresholds, meaning that people don’t taste or smell the compounds in water until a relatively high concentration. Chlorinated solvents have taste thresholds around several hundred ug/L; MTBE is at least one if not two orders of magnitude less. Taste thresholds are highly dependent on the individual.
Fate and Transport
DNAPL transport has been studied in three categories, in order of decreasing scientific understanding: sand and gravel, fractured clay and fractured rock. Research on DNAPL fate and transport in fractured rock is nascent and appears to be dominated by Beth Parker and John Cherry at the University of Guelph (but of course I learned this at a course taught by Cherry).
Sand and Gravel
A series of field experiments were conducted in the fine sand aquifer of Canadian Forces Base Borden (Ontario), starting in 1989, to explore the transport of different chlorinated DNAPLs with different release scenarios. This work isn’t news anymore, so I won’t go into more than to say that the DNAPL traveled deeper than expected and along very thin, slightly coarser-grained beds in an fairly homogeneous sand, forming multiple layers of free phase product at discrete depths.
The extent of the fractures in an aquitard are the most relevant bit of information: do fractures extend through the unit and connect to an underlying aquifer or not? The importance of aquitard fracture extent was discovered by accident during a controlled release experiment of PCE at Base Borden in 1991 (I can’t find a publication of this specific experiment, though I believe it became part of Parker’s PhD dissertation). A falling head hydraulic test indicated that the aquitard was a sufficient barrier to flow; however, the hydraulic test failed to indicate slight leakage, which DNAPL is happy to exploit. It’s easy to forget that aquitards are defined in terms of water supply, not contaminant transport, and methods to test the aquitard’s hydraulic properties are likely insufficient to determine contaminant transport properties. How to determine that an aquitard has continuous fractures? Carefully inspect cores.
There is still very little literature on DNAPL in fractured rock. There are several references addressing fluid flow in fractured rock (conference proceedings, National Resource Council “Rock Fractures and Fluid Flow” and a guide to regional groundwater flow available as a PDF here). Modeling demonstrated, before any field experiments were conducted, that the “orderly” and interconnected fracture network in sedimentary rock would generate a dispersion-dominated plume. In the 1997, a major field effort was initiated at Santa Susana (Simi, California), to study the fate of large volumes of TCE, which had been used to clean rocket test components on top of a sizeable shale/sandstone hill. Cherry spent a lot of time on the methods (message: don’t open a hole that becomes a transport pathway) but the surprising discovery at this site was that despite the high water table beneath the source area, TCE was not appearing at discharge locations. Analysis of numerous rock cores indicated that transport of DNAPL was dominated by matrix diffusion, not advection, and diffusion had sufficiently retarded TCE such that it never appeared at discharge points in the valley below. After four decades, no DNAPL source zone remained. (Parker presented this at the Battelle Conference on Chlorinated and Recalcitrant Compounds, Monterey, CA, May 24-27, 2010.)
If you made it this far, thank you! This is my first real blog post. I encourage you to leave comments or questions. Were you hoping that some nugget of DNAPL properties would be addressed here that wasn’t? Let me know and I’ll do my best. If you have suggestions on how I could write better, I would really appreciate those comments. If you don’t want to leave a comment in this public forum, feel free to email me or DM me on Twitter. Cheers.