The objective of this project was to demonstrate the use of a mobile unit containing a cone penetrometer system coupled with two types of laser-induced breakdown spectroscopy (LIBS) sensors or an x-ray fluorescence (XRF) sensor for real-time semi-quantitative analysis of heavy metal distribution throughout a soil profile.

The LIBS and XRF sensors are mounted on a SCAPS mobile cone penetrometer probe, which can be pushed to the depth of interest. The sensors are then advanced to the next depth without retracting the probe. The two LIBS systems, Fiber-Optic LIBS (FO-LIBS) and Downhole-Laser LIBS (DL-LIBS), operate by generating a laser-induced spark at the probe-sample interface. The spark results in rapid heating, vaporization, and ionization of the soil within a small volume around the focal point of the laser. The difference between the LIBS sensors is that the FO-LIBS produces the laser energy in the truck while the DL-LIBS produces the laser energy within the probe. The resulting plasma is spectroscopically analyzed. The XRF system emits x-rays that excite a core electron, which then decays to emit a characteristic fluorescence x-ray that is measured in a truck-mounted detector.

Demonstrations were conducted at Lake City Army Ammunition Plant (Independence, MO), Naval Air Station (North Island, CA), Hunters Point Shipyard (San Francisco, CA), and Camp Keller (Biloxi, MS). The sites' soils ranged from silt loams to sands and were contaminated with a suite of heavy metals. Results from the three sensors were compared semi-quantitatively with inductively coupled plasma (ICP) laboratory analysis on split and spiked samples. There was 97 percent agreement between the LIBS methods and the ICP and 91 percent agreement between the XRF and ICP. These results support the use of the SCAPS heavy metal sensor system for real-time soil screening. Also, sensor-derived heavy metal point estimates enable a rapid three-dimensional contaminant representation in different soil matrices. To note, increased moisture and matrix effects limit this technology in the saturated zone. The average cost of operating the SCAPS truck and 4-person crew was approximately $6,380 for a 10 hour day, which included an equipment rate of $3,700 per day. These costs would increase in difficult geologic soils (i.e., cobbles).

The SCAPS Heavy Metal Sensors were capable of providing a real-time semi-quantitative determination of soil heavy metals. Cost savings represented by this technology are achieved through rapid, site-wide screening for the purpose of locating sites for more detailed analysis. Unit costs are estimated at $7.12 per sample for 1,800 measured samples as compared to $512 per sample for 60 samples collected by conventional drilling and off-site analysis and $172 per sample for 60 samples collected by direct-push samples and off-site analysis. In addition, sensor-derived point sample analysis (as opposed to contaminant averages per depth interval using traditional sampling techniques) enables a more rapid and cost-effective analysis of contaminant distribution within a soil profile. (Project Completed - 2003)