1. Environmental Sciences Division and 2. Instrumentation and Controls Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831
3. Special Technologies Laboratory, Santa Barbara, CA
4. National Security Program Office, Lockheed Martin Energy Systems, Oak Ridge, TN* corresponding author
Environmental Sciences Division
Oak Ridge National Laboratory
MS-6036
Oak Ridge, TN 37831, USA
Tel: 423-574-7321
Fax: 423-576-8543
E-mail: RVX@ORNL.GOV
(Excerpts Reproduced with
Permission of Author.
This page is part of the Humanitarian Demining Web
Site at the University of Western Australia)
Wide area detection of unexploded ordnance (UXO) is difficult to achieve. Most techniques for UXO detection depend on slow and laborious probing of small areas; this is time-consuming and expensive. We are developing a method for examining large (hundreds of acres) areas of land in a few hours. The technique utilizes a strain of bacteria that responds to trinitrotoluene (TNT) and which has been genetically engineered to fluoresce brightly in its presence. The bacteria are sprayed over the area under investigation and allowed to contact the soil. Fluorescence detection, either electronically or visually, is already faster than all other detection methods, and we are developing methods to examine the area from an airborne platform, thus making detection extremely efficient. This technology has been demonstrated at a field site in South Carolina, at which all five targets (ranging from three ounces to ten pounds of TNT) within a quarter acre tract were found. The bacteria are easy to grow and to transport, and might be used on a large variety of land surfaces and under a range of climactic conditions. Detection equipment ranges from a versatile laser-induced fluorescence to the relatively simple and inexpensive handheld method, which is ideal for densely overgrown areas.
Schematic diagram of process for detecting TNT concentrations in the environment with bio-reporter bacteria.
Resulting scan from 4 metre wide by 3 meter long (high) test area - mine targets are within 1 meter of indications. There are false positives, apparently associated with water drainage. "In this figure the positive signals are associated with plant material on the site. In fact, the shape of blades of grass can clearly be seen. We are uncertain of the reason for the close association between our bacteria and the plant material. It may be due to the waxy cuticle on the leaves which allows the bacteria to pool there. It may be due to transpiration of TNT through the roots and stems of the plants to give a higher concentration on the leaves. It may be that the plant is converting the TNT to another compound that works even better with the bacterial strain. These hypotheses are the subject of continued investigation."
Although we have only conducted a single field test of this technology, the
results of that test were so favorable that we believe full commercialization
of the technology can soon be realized. A second field test of the technology
is scheduled for the Fall of 1999, and during that test we will vary some detection
parameters in order to increase our sensitivity, especially at higher altitudes.
We believe that mounting a detector, such as the LIFI unit, on an airborne platform
is practical. When finished, the system should allow a sweep across a broad
expanse of ground at a slow flying speed, probably using a helicopter. In this
manner hundreds of acres could be examined in a single day.
There are, however, many basic research questions that must be addressed before
this system can be commercialized. We have not produced a bacterial strain that
detects RDX, the other explosive commonly used. Producing such a bacterial construction
is a priority for ORNL. It is also noted that the bacterial solution soaked
into the ground when it was sprayed on, due to the dryness of the soil. A means
to hold the bacteria on the soil surface is therefore required. For this reason
the interaction of the bacteria with the plant surfaces is especially welcome,
although we must still study this interaction in order to optimize the conditions
under which it functions. Understanding the mechanism of interaction is critical
to the success of this technology, as well as identifying the plants that will
work the best. Our greatest challenge will undoubtedly come in desert environments,
where the soil is exceptionally dry and there is little plant life. Unfortunately
this is also where the majority of the world’s landmines (and other UXO)
are buried. We are confident that a means of applying the MMDS technology will
be found that will function well even in such a hostile environment.
Finally, the detection technology, while showing excellent specificity in the first test, can still be improved. It should be practical from a higher altitude (thus improving safety) and at a faster scanning speed. For ground use, especially in heavy jungle growth, the simpler optical detection system will be ideal, although even this system can be improved through the use of inexpensive electronic detectors rather than trusting the skill of the individual operator.
In its commercialized format, MMDS should be able to cover hundreds of acres at a single time and accurately map sites containing UXO of all types. The only exception to this detection scheme is for munitions that are so well constructed that neither impact nor time was sufficient to crack the ordnance casing. For even the best-constructed munitions, corrosion over a period of years is expected to allow escape of sufficient concentrations of explosive to enable detection.
The professional assistance of Mr. Keith Williams, NEWTEC, and Maj. Mike Keleher, U.S. Army (ret.) is gratefully acknowledged. Research sponsored by the Defense Threat Reduction Agency, U.S. Department of Defense. Oak Ridge National Laboratory is managed by Lockheed Martin Energy Research Corp. for the U.S. Department of Energy under contract number DE-AC05-96OR22464.