The US Army Engineer Waterways Experiment Station (Vicksburg, MS) has developed an in situ screening system for identifying and quantifying subsurface heavy-metal soil contamination. A three-dimensional (3-D) map is created that estimates and identifies the plume shape and extent of metal contamination in a specified area. The system is based on the site characterization and analysis penetrometer system (SCAPS) and its laser-induced breakdown spectroscopy (LIBS) sensor probe.
Brian Miles, principal investigator for the project, says that the Army is interested in detecting heavy metals such as chromium, lead, cadmium, and mercury, which may contaminate Department of Defense (DoD) properties. "It is important to utilize a subsurface profiling system such as SCAPS to delineate the underlying contamination plume shape and extent in order to aid regulators and remediation experts in their cleanup efforts. The extent of the surface contamination merely defines the intersection of the contaminant plume and the surface, but yields little information on the shape and size of the contamination plume below the surface," Miles says.
Although the cone penetrometer system is not new and LIBS is not a novel technique, the combination of LIBS technology with the SCAPS system is new and allows rapid subsurface metals analysis with no sample preparation. The SCAPS platform consists of a custom-built truck with hydraulic ram capable of a 20-ton push force. The ram pushes pipe sections of the penetrometer into the ground. Additional sensor probes and associated electronics complete the system.
The LIBS sensor probe is housed in the first pipe section. The probe includes the laser, focusing, and collection optics, and soil geophysical sensors for soil classification. In this LIBS application, a single 80-mJ pulse from a Q-switched Nd:YAG laser provides flux levels sufficient to rapidly heat, vaporize, and atomize a small volume of the soil. The laser pulse further heats the vapor to a plasma with a temperature on the order of 5000 K. Immediately after plasma formation, the emission spectra consist primarily of broadband blackbody radiation from the hot plasma. As the plasma cools, the blackbody component decays and the photons formed by downward atomic transitions dominate.
The spectral content of the plasma radiation is analyzed by atomic emission spectroscopy (AES). In AES, the wavelength of a spectral peak indicates which element is present. The strength of the peak is typically proportional to the amount of an element in the sample.
A penetrometer push begins as the sensor pipe is forced into the ground. A hollow push pipe, 1 m long and with a 4.45-cm outer diameter, is slid over the sensor pipe`s trailing umbilical cord and screwed to the top of the sensor pipe. As the sensor pipe continues downward, additional 1-m-long push pipes are added until the desired depth is reached. The soil geophysical sensor provides soil classification as the pipe is pushed.
Once the desired depth is reached, the probe begins retraction, leaving a sacrificial sleeve in the bottom of the hole. The laser starts firing through the now-exposed window in the side of the probe. A fiber positioned a short distance from the plasma collects a portion of the plasma light, conveys it up the hole to the truck, and introduces it into a spectrometer with a gateable intensified charged-coupled-device array detector. The system currently samples the soil every 0.87 in. of upward travel. After the retraction is complete, grout is poured into the hole to prevent cross-layer contamination.
Mapping the plume
The distribution of contamination is determined by feeding data from individual probe pushes into an image-processing algorithm developed at the Waterways Experiment Station. A plot of the data as function of depth and intensity creates a 3-D model of metal peak strength, which, after accounting for previously mentioned effects, can yield numerical values of subsurface contamination.
The strength of the AES spectral peak does not correspond directly to the amount of contamination. Soil moisture, grain size, and other factors will affect peak strength. If the soil is wet where lead contamination is being measured, for example, the spectral peak will be weaker than in dry soil because some of the energy from the laser pulse is evaporating soil moisture rather than creating plasma. Eventually, the US Army Corps of Engineers hopes additional soil sensors will remove the influences of soil moisture, grain size, and other effects.