Badger Army Ammunition Plant (BAAAP) in Sauk County, Wisconsin, is one of many inactive Army ammunition plants currently under the control of the Department of Defense (DoD) with transitioning missions in place. These plants are in varying stages of transfer out of DoD control. In order to transfer these properties, DoD must characterize and decontaminate the properties to a level protective of human health and the environment. To accomplish this task, many buildings used in the production, loading, handling, and storage of explosives will have to be demolished or characterized and decontaminated. BAAAP alone has more than 1,400 buildings on the installation that have to be addressed. The contaminants of concern associated with the buildings at BAAAP include nitrocellulose (NC), nitroglycerine (NG), dinitrotoluene (DNT), and common environmental compounds such as asbestos-containing material (ACM), solvents, and metals.

A previous ESTCP demonstration was performed between April 28, 2002, and May 13, 2002, to demonstrate candidate field test methods for NC and NG. The methods evaluated were Raman spectroscopy, EXPRAY™ colorimetric indicator, and the Cold Regions Research and Engineering Laboratory (CRREL) Royal Demolition Xplosive (RDX) colorimetric field screening method. The methods were used to test for the presence and/or concentration of NC or NG in soil samples and concrete slabs. Attempts were made to compare the results from these field measurements to laboratory analyses of NC and NG in the same materials to evaluate the reliability of the field screening and analytical methods for identifying and quantifying NC and NG in buildings and soils. The results and lessons learned were presented in the Phase I Final Report. Due to a number of factors, including the lack of energetic compounds in the buildings used for the demonstration, attempts at validation of the field methods for detection of these materials were inconclusive.

This ESTCP demonstration targeted areas with higher suspected explosives contamination to obtain samples containing measurable amounts for testing.


The objectives of this ESTCP demonstration were to evaluate and document the performance of three candidate experimental field analytical methods for detecting and quantifying NC and NG associated with structural concrete pads, underlying soils, and structural building materials such as framing timbers. The technologies evaluated in the field demonstration were DROPEXPlus/EXPRAY™ colorimetric indicators, gas chromatography/thermionic ionization detection (GC/TID), and the CRREL RDX colorimetric field screening method.

Demonstration Results

DROPEXPlus and EXPRAY™ tests were found to be useful tools for screening the presence of significant concentrations of NC and/or NG in the field or on sample extracts. In combination with other field methods, they are beneficial screening tools for identifying areas that contain explosive contamination in buildings above specified limits. Detectable levels are matrix-dependent, with low confidence in results at or near the detection limit. The CRREL RDX method of analysis gave a relatively low response for NC compared to NG, and the response was easily impacted by matrix interferences. Modifications to the CRREL RDX procedure greatly increased the method response for NC and retained the response for NG.

The increase in response made the modified method more robust for NC analysis and allowed analysis of NC on the three sample matrices. Neither CRREL method is specific to NG or NC, and both provide a response to the total of NG and NC. Tests conducted with NC on concrete using the CRREL methodology showed that recovery of NC from the matrix was a function of both time and the manner in which NC was deposited on the matrix due to decomposition of NC by the concrete matrix. The degradation of NC was the same effect noted in the bench test portion of testing for NG. The instability of NC/NG compounds on concrete matrix makes analysis difficult not only because of the potential impact on samples during handling and preparation, but also because of the effect on matrix standards.

Compound detection performance and concentration metrics for NC/NG using the CRREL methods compared to laboratory reference method results were not met for the three matrices. This was due to a combination of factors, including insufficient positive result data for comparison on the concrete matrix; nonhomogeneity of the soil matrix due to pieces of propellant material in the samples; and in general what was felt to be highly biased and false positive results from the reference method for NC analysis on all matrices. GC/TID analysis of NG on soil, wood, and concrete samples was a sensitive and reliable method providing results consistent with the laboratory reference Method 8330, but the comparison of concentration results did not meet performance metrics in the field demonstration. This was attributed to insufficient positive result data on the concrete and soil matrices and matrix interference with the wood sample analyses. Both methods may be subject to matrix interference effects, and quality control (QC) samples should be included to help assess data quality. The bench-scale study demonstrated good GC/TID results agreement with NG spike concentrations on clean background matrix samples.

Implementation Issues

Field analytical methods for detecting and quantifying NC and NG associated with DoD buildings and underlying soils represent additional tools that will aid in the decision-making process for transfer of property at BAAAP. Successful demonstration of these methods will help streamline the property disposal process.