A child innocently walks in an ordinary looking field where a small object lurks beneath the ground. Her foot lands on the small container filled with a powerful explosive. Instantly, she is blinded by a bright flash, her ears ringing from the loud metallic crack of a land mine she has triggered. Then she screams in pain, her body tossed by the blast, seared by the intense heat. The explosion propels fragments of the casing, pieces of her shoe, and bits of her shattered bones into her leg. She probably will survive - these devices are meant to hurt, not to kill. But she likely will lose a limb, or two.
Human catastrophes like this occur every day in places where antipersonnel land mines have been planted by opposing factions in armed conflict. According to the International Committee of the Red Cross, in Geneva, land mines around the world claim a victim every 20 minutes.
Long after the combatants have made their peace, these concealed abominations continue to terrorize people struggling to return to normal lives. They threaten peacekeeping forces and relief workers whose labors are crucial to a war-torn region's return to normalcy.
Two solutions are required. One is to quickly and cost effectively remove and destroy the innumerable antipersonnel land mines already in place worldwide. Another is to ban the production and use of the stealthy and indiscriminate killers to prevent their further emplacement. The first requires technological innovations; the second, political will.
Finding, removing, and destroying land mines in the aftermath of armed conflict is a prerequisite to a return to normalcy. It is a dangerous, tedious chore, and R&D efforts now are being focused on technologies that will enable the job to be done more quickly, more safely, and less expensively. In particular, researchers are looking at detection methods that specifically pinpoint land mines.
Attempts at a political solution have been highly visible recently. Moves toward an international treaty to ban antipersonnel land mines were widely discussed in January, when the Clinton Administration announced it would negotiate such a treaty through the United Nations Conference on Disarmament in Geneva rather than through a Canadian-led effort called the Ottawa Conference.
That announcement came eight months after the Clinton Administration declared its policy on antipersonnel land mines: aggressive pursuit of an international agreement on a ban as soon as possible. At the same time, the Administration is reserving the right to continue using the weapons in the Korean Peninsula until alternatives are found or the risk of aggression in that area is removed.
The policy commits the U.S. to destroying its stockpile of non-self-destructing land mines - dumb land mines - by 1999, except for what will be used for training and for defense at the Korean Demilitarized Zone. But until an international treaty is in effect, the U.S. reserves the right to use smart mines - those that become inoperative after some period of time.
Parties pressing for an immediate ban were disappointed. The policy has been widely regarded as a less-than-total commitment to a treaty banning land mines. And the January decision to take the Geneva route was interpreted as a delaying tactic.
In a statement, Sen. Patrick J. Leahy (D-Vt.) said he thinks the Geneva Conference on Disarmament is not the best route to a total ban. Leahy is widely regarded as one of the principal catalysts of the movement toward a global ban on land mines. He has called attention to the issue by initiating laws to ban U.S. exports and use of antipersonnel mines, establishing a fund in the U.S. foreign aid budget to assist victims, and launching in Congress and the United Nations a campaign for a ban.
"The [Conference on Disarmament] process requires step-by-step consensus that rewards holdout states, who effectively have a veto that retards or prevents strong agreements," Leahy said. "I am skeptical [it] will move quickly toward the goal of banning land mines."
Nongovernmental organizations (NGOs) also were disappointed by President Clinton's move. They believe the decision is a signal that the U.S. does not want to see rapid progress toward a ban.
The Conference on Disarmament has not yet even decided whether it will put land mines on its agenda, explains Mary Wareham, coordinator of the U.S. Campaign To Ban Land Mines, based in Washington, D.C. The conference "may want to put other issues before land mines," she explains. But unlike chemical, nuclear, or biological weapons, land mines already are killing people every day, she continues. "The issue cannot afford to wait for years to be negotiated."
Another drawback of the Conference on Disarmament is its exclusivity, says Stephen Goose, program director of the Human Rights Watch Arms Project, also based in Washington. It does not have good representation of the countries that most need a treaty to ban mines - those most contaminated by land mines.
The alternative Ottawa Conference is supported by NGOs. This Canadian initiative aims to have a draft treaty by year's end and is open to any state willing to commit to a total ban on land mines. The Clinton Administration says it welcomes the efforts of the Ottawa process but has elected not to participate, in part because the Canadian initiative lacks the support of China and Russia, two major producers of land mines.
Goose believes those two recalcitrant states would feel pressure to commit to a ban when they see dozens of nations coming to Canada to sign a comprehensive ban treaty. "The Chinas and Russias of the world need to see that the international community objects to these weapons of war," he says.
Significantly, says Goose, the developing world - especially those countries that have been the biggest users of antipersonnel land mines and the ones that are most affected by them - are getting behind the Ottawa process. At this stage, he adds, it may be more important to get nations such as Angola, Mozambique, and Cambodia to commit themselves to not using land mines than to get China and Russia to agree to cease producing them.
Use of antipersonnel land mines is regulated by general humanitarian law, which prohibits attacks on civilians. It is also restricted by the second protocol of the 1980 treaty widely referred to as the Convention on Conventional Weapons.
Last May, some changes to the protocol on land mines were adopted in response to public outcry over the crisis. The amendments required, among others, that all antipersonnel land mines be detectable with commonly available technology.
The new provisions are only slight improvements and came very slowly, Leahy noted last December at a conference held in Washington, D.C., on land mine neutralization and removal. A few countries used their veto to ensure that even the modest changes would not take effect for many years, he added.
Negotiating a ban might be hastened if the meetings were held in the middle of a minefield in the Cambodian countryside, Leahy suggested last December. If at the end of the day the negotiators had not reached agreement, he proposed, they'd then pick which field to meet in the next day. "I suspect the immediacy of the problem would become more evident than otherwise."
Better technology to find and remove the mines is urgently needed, given the dimensions of the problem: The International Committee of the Red Cross estimates that 120 million unexploded land mines are scattered throughout about 70 countries. The UN estimates that 80,000 were cleared in 1995; during the same period, 2.5 million more were installed.
"The clearers cannot keep pace with the sowers," says Harry N. (Hap) Hambric, project leader of the Humanitarian Demining Technology Development Program of the Department of Defense. "More R&D is required if the clearers are to win."
The cost of clearing is way out of proportion to the cost of the weapon. Average prices for antipersonnel mines vary from $3.00 to $15 per unit, according to the International Committee of the Red Cross. By contrast, clearing can cost as much as $1,000 per unit.
Postconflict clearing of land mines, called humanitarian demining, is painfully slow. In Cambodia, for example, 4 million to 6 million land mines were strewn between 1970 and 1990 during internal power struggles among various armed factions. Only 4.6 square miles can be cleared per year with current technology, according to Sam Sotha, director of the Cambodian Mine Action Center. At this rate, it will take almost 100 years to clean up Cambodia alone.
The UN estimates it will take $33 billion and 1,100 years to clear all the mined areas in the world with current technology. "An impossible sum and a totally unacceptable period of time," said Leahy last December." We have to do better," he said. "I'm optimistic that with research we can cut the cost and time by factors of up to 100."
Humanitarian demining is slow and expensive because the methods date back to World War II. Also, its objectives are different from those of tactical countermining, for which effective technologies exist.
During armed conflict, tactical countermining comprises operations that allow an attacking force to penetrate or avoid mines rapidly as it attacks a target or to speedily clear areas to sustain specific operations. The pace is very quick. Individual mines need not be found, and casualties from mines and other weapons are expected and accepted.
In humanitarian demining, however, the goal is to detect and then remove or destroy each land mine. Safety is of utmost importance, and casualties are unacceptable.
Clearers manually search with a metal detector and a pointed stick and, like archaeologists on a dig, extract the mines with small digging implements. When the detector locates a metallic object, the clearer carefully probes the earth. Upon hitting something hard, the operator then "feels" the contour of the object to figure out what it is. If it's a mine, it is exposed, coaxed painstakingly out of the ground, and neutralized by detonation or by removal of the triggering mechanism. Says Harvard University physics professor Paul Horowitz, who led a study on new technological approaches to demining:" The process is very tedious and unforgiving of carelessness."
Even before the clearers can do their work, a major hurdle is getting to the mines themselves. "Think about Cambodia," says Hambric. The mines were placed in farming fields. Today, those fields have brush and trees. "You have to clear out the vegetation before you can start looking for the mines," he explains.
Clearing is also complicated by the variety of land mines, despite the simplicity of the underlying concept: A sensor detects a target. The stimulus causes a small amount of sensitive detonator to explode, which in turn sets off a larger quantity of the less sensitive explosive in the main charge. All the components are neatly encased in metal, plastic, or wood. The devil is in the details, resulting in some 2,500 mine and "fuse" combinations, according to William C. Schneck Jr., a project engineer with the Army's Countermine Division in Fort Belvoir, Va.
The sensor and the activating mechanism together make up the fuse, which can be configured in many ways. Fuse technology has evolved from simple mechanical pressure sensors to sophisticated electronic processors that may even control several mines at a time, explains Schneck. Electronic sensors that can analyze a wide range of signals from their targets - acoustic, seismic, magnetic, or thermal - have become available.
The favorite explosive for the main charge is TNT (2,4,6-trinitrotoluene). Other explosives used are RDX (1,3,5-trinitro-1,3,5-triazacyclohexane); composition B (a combination of RDX, TNT, and wax); tetryl (2,4,6-trinitrophenylmethylnitramine); and C-4 [a combination of RDX, polyisobutylene, di(2-ethylhexyl) sebacate, and fuel oil]. In addition, mines may have a booster charge to enhance the power released by the detonator to a level that's enough to initiate the main charge.
What happens when a land mine explodes is also variable. Blast mines burst on contact, attacking a single individual. These mines are small and usually have very low metal content, says Schneck.
Other mines threaten a greater number of targets. Bounding mines leap to a height of up to 2 meters before blowing up, spreading mayhem as far as 30 meters in all directions. Many such mines, particularly those produced by the former Soviet Union, have combinations of sensors in their fuses. Directional fragmenting mines erupt in a spray of shrapnel, propelling deadly fragments in an arc that reaches as far as 200 meters.
Not only are land mines variable, but the environments where they have been placed - terrain, vegetation, and climate, for instance - also are variable. No single detection device exists that can address all these permutations. "Lots of people think in terms of a silver bullet," says Hambric. "That's unrealistic in humanitarian demining. What we need is a tool box."
That tool box will include devices of all sizes, shapes, and levels of sophistication - ranging from rugged vehicles that would cut through a field and simply blast every mine in their paths to miniature robots that can gingerly locate mines without setting them off.
Indeed, a tool box is taking shape at the Army's Countermine Division, where Hambric oversees the development of technologies for humanitarian demining. Already, the tool box has vehicle-mounted detectors equipped with special cameras, remotely operated mine clearers, chemicals and methods to neutralize mines, and devices to clear vegetation, among others.
"We're developing prototypes, identifying and leveraging existing equipment with potential for low-risk success, testing them in the U.S. and in places like Bosnia, evaluating them, and modifying them until they can be handed over to the private sector," says Hambric.
The tools being developed by Hambric's group are available free to companies that wish to commercialize them." We'll give it to anybody who wants them," says Hambric, adding that DOD is not out to make money off the technology development program. Furthermore, DOD insists that any technology from the program can be transferred to other countries.
With $14.7 million in fiscal 1997 for research and development, the Humanitarian Demining Technology Development Program is eager to add more tools to its box. Last September, it solicited proposals to develop new technologies, with emphasis on those that can be tested and delivered within 14 months.
"We're not looking for high-tech long- term solutions down the road," says Hambric. "We try to do things that can cure things right now; we can't wait too long."
Hambric's sense of urgency is shared by many. For example, Richard M. Walden says, "Scientists tend to take too long to get stuff going." Walden is president of Operation USA, a Los Angeles-based international relief organization committed to a global ban on antipersonnel land mines. The group is scouring the country's national laboratories and research agencies for advanced technology that can be applied to the land mine problem.
"We think the science already exists to find antipersonnel land mines, but it's being used elsewhere," he says. Scientists need to think about whether anything they're doing can be used for land mines, he continues. "We need people to jump track and apply their science to our problem."
The idea of jumping track came to Walden during a 1995 meeting with robotics scientists and engineers at the National Aeronautics & Space Administration's Far West Regional Technology Transfer Center in Los Angeles. "I told them about land mines we can't find with metal detectors because they're made of plastic," he recalls.
The response from the NASA scientists and engineers was very interesting, says Walden. "Someone from the Jet Propulsion Laboratory said, 'That's the most ridiculous thing I've ever heard ... when we're getting ready to send a robotic probe to Mars that, among other things, will be shooting ground-penetrating radars, picking up rocks, analyzing them for content, and then e-mailing the results back to Earth,'" he recalls.
Lack of interaction among government agencies is part of the problem, says Walden. As an example, he says, prior to his contacts with NASA scientists" they had not been approached by DOD or anybody else about applying their technology to demining." He also mentions the "intelligence community" and the chemical industry as not engaged in the search for solutions.
The National Security Agency, the National Reconnaissance Office, and the Central Intelligence Agency, which have expertise in sensing devices, are not participating, says Walden. He adds: "The chemical industry is not involved at all. There's a whole area of sniffers that probably exist as security devices in chemical factories just to alert people of dangerous odors or spills. The same devices could be very useful in sniffing mines."
For their part, NASA scientists believe they have robotics and related technologies developed through the space program that can be applied to land mines. Humanitarian demining has obvious analogies to planetary exploration, says Karen G. Engelhardt, formerly manager of robotics and health care projects at NASA's technology transfer center in Los Angeles. On Mars, for example, as on a minefield, robots must be able to dexterously navigate, sense various agents, and manipulate objects in unstructured environments. According to Engelhardt, NASA also is developing a new generation of robots for subsurface exploration, which could be applied to digging up land mines.
Among NASA's gizmos are so-called nanorobots. These are miniature robots - about 15 cm long and weighing between 10 and 100 g - that could do different things and collectively be used to "colonize" unexplored areas. Says Engelhardt: "We might think of cockroach or termite types that can burrow and approach land mines and not trigger them because these robots are so lightweight."
Right now, NASA has robots that can be evaluated for deployment in minefields, says Engelhardt, who is now president of KG Robotics, Los Angeles. However, the technologies are not yet in the hands of those who need them and can evaluate them. A wide technology gap exists, she says, because the people doing the research and the companies eager to take new technologies to the market do not always know how the transfer can occur." What we need most is partnerships among NGOs, the business sector, researchers, and the leading-edge labs."
Robots may be extremely versatile and flexible, but ultimately they will be limited by the component that detects the mines. Mine detection has been dominated by technologies that exploit features associated with the threat. Sensors such as metal detectors, ground-penetrating radar, and infrared imagers detect aberrations in the electrical, optical, or thermal properties of an area caused by the presence of land mines.
Because those detectors seek anomalies rather than cues representing the threat itself, they inherently cause a lot of false alarms, says Regina E. Dugan, a program manager at the Defense Advanced Research Projects Agency (DARPA), Arlington, Va. Rocks, vegetation, soil, and even shadows can cause discontinuities in physical properties, Dugan explains. As for metals, "they're everywhere that humans have been," she adds. Furthermore, clutter is inversely proportional to the size of the target. As the size of the mine goes down, the sources of clutter increase dramatically.
With the drawbacks of current land mine detectors, it is not surprising that a DARPA-commissioned study led by Harvard's Horowitz concludes: "The greater problem isn't finding mines, it is discriminating candidate mines from the thousands of pieces of metallic 'clutter,' much of it the detritus of war and artifacts of civilization."
For sensors to be able to discriminate, they must detect what is unique to the threat, such as chemicals associated with the land mine, particularly the explosive itself.
The DARPA-commissioned study," New Technological Approaches To Humanitarian Demining," examined discrimination technologies that can be used in humanitarian demining. Some approaches, such as nuclear quadrupole resonance, depend on probing the bulk of the explosive. Others are based on vapors or traces of chemicals exuded by the mines. The suggestions range from the labor-intensive, low-tech but highly effective use of dogs, to relatively high-tech approaches based on mass spectrometry, to the somewhat exotic notion of deploying fruit flies and bees.
With a deeper understanding of the demining problem, chemists and chemical engineers should be able to contribute to demining technologies, says Divyakant L. Patel, a physical scientist with DOD's humanitarian demining program. Designing sensors to detect chemical traces from land mines presents a special problem, he explains. Time is not so critical, because humanitarian demining requires all mines be found. But sensitivity at very low levels of analytes is crucial.
The amounts of explosives in land mines available for detection are extremely small, in the sub-parts-per-trillion level, says Patel. To begin with, the explosives usually found in land mines have very low volatilities. Then, as the molecules make their way from inside the casing to the soil to above the ground, their numbers decrease as some are trapped in soil particles, dissolve in soil moisture, or are eaten by bacteria. Depending on weather conditions, molecules that emerge can be degraded by sunlight, blown by wind, or washed away by rain.
With the advances in chemically specific sensors and even biosensors, detecting trace amounts of specific chemicals in a lab is straightforward, says Patel. But large lab systems that require highly trained operators cannot be used in the field. "We need rugged systems that are affordable, portable, and operable by people with minimum training," he says.
One of the major problems in designing chemical sensors for land mines is how to transfer the traces of explosive chemicals from the soil to the instrument. Sampling methods must take into account TNT's unique characteristics - especially its stickiness. It sticks on all surfaces, even on nonstick polymers such as DuPont's Teflon. The only exceptions are gold and indium, says Patel.
Chemists also must explore ways to concentrate the minute amounts of explosive molecules above ground, to detect them in the soil, or to increase the amounts that emerge from the soil, says Patel. "We need suction or microextraction technologies that won't trigger the mine," says Patel. Absorbents with selectivity and affinity for TNT and systems that will heat the mine or loosen the surrounding soil to cause more molecules to volatilize will be useful, he adds.
An alternative to directly detecting the main explosive chemicals is to detect their associated impurities. For example, says Patel, instead of sensing TNT, one can try to find its more volatile impurity 2,4-dinitrotoluene (2,4-DNT).
Chemists can also try to develop reactions to make the explosive molecules visible, says Patel. For example, a reagent that can be sprayed on the soil that will visibly change after reacting with traces of TNT will be helpful in determining whether a given area is dangerous.
Given the challenges of chemical detection, it is not surprising that, so far, the most successful chemical detectors for land mines are dogs. Dogs are extremely sensitive and "exquisitely good at finding land mines, outperforming any physical systems currently available,"says DARPA's Dugan. They can be trained to sniff a wide range of explosives, and they can easily move about in various terrains.
But dogs get tired. Illness can reduce their reliability, and they need the careful attention of a human partner. There also is the risk of dogs accidentally triggering the land mines.
The effectiveness of dogs has caught the eye of DARPA. As DOD's central R&D organization, DARPA's mission is to help maintain U.S. technological superiority by developing high-risk, high-payoff ideas. The agency now is embarking on an aggressive $25 million, three-year program to develop technology that could lead to an electronic analog of the canine nose.
Chemical detection should be fast, highly specific, and highly sensitive, says Dugan, who is managing DARPA's canine nose program. Any one technique currently available meets only two of those three goals, she explains. Furthermore, traditional techniques focus on the detection of individual chemicals. The mammalian nose, on the other hand, is capable of identifying complex odors.
Dugan explains that humans, for example, don't identify specific chemicals when they whiff the aroma of red wine, which is due to hundreds of chemicals." Similarly, if we are to develop an electronic canine nose, odor recognition is key," she says.
Creating a dog's nose is an old goal, says Dugan, but only recently has it come within reach, with the advances in the understanding of mammalian olfaction (C&EN, Dec. 23, 1996, page 18). Part of the goals of DARPA's canine nose program is to answer basic questions about olfaction that are central to sensor development. For example, how the dog's exceptional sensitivity comes about is not yet known. It is not even clear what chemical cues make up the" suite" of signals that the dog interprets as the smell of a land mine.
"It's a technology with enormous potential, and the investigators are so torqued up," says Dugan, a mechanical engineer. Part of the enthusiasm comes from the interdisciplinary approach - basic researchers studying the neurobiology of olfaction teaming up with widget makers, she explains. Part of it comes from being involved in a project that could not only revolutionize chemical sensing but also lead to a device that helps solve a worldwide military and humanitarian problem.
"We're going for the big payoff," says Dugan. "Maybe we won't achieve it in three years. But I'll be damned if we're not going to try."
One group that's already trying to emulate the canine nose is led by chemistry professor Nathan S. Lewis at California Institute of Technology. Lewis is trying to use nonspecific detectors for land mine detection. Instead of responding to individual chemicals, these detectors recognize patterns generated by an array of sensors interacting with a vapor.
The sensors are composites made from swellable organic polymers. The polymers swell in response to chemicals sharing unique properties rather than to specific chemical species. The polymers contain a sprinkling of conducting particles, and when they swell in the presence of an odor, they produce a signal, a change in electrical resistance.
The swelling of individual polymers in the presence of a vapor is translated to an electrical fingerprint unique for that vapor over the array of sensors. The more sensors there are in the array, the greater the discriminatory power. Lewis says his group is working on putting thousands of such sensors on a chip. The "nose-on-a-chip" research is part of a $6 million multidisciplinary university initiative in land mine research that is funded by the Army.
A similar technology is used by AromaScan, a manufacturer of odor-sensing instruments based in Crewe, England. The company's electronic nose, called AromaScanner, consists of an array of 32 gas sensors made from electrically conducting polymers (C&EN, Dec. 18, 1995, page 30). The polymer conductivities change when exposed to vapors, producing a fingerprint that is displayed in three dimensions with a computer. The company says preliminary work with explosives has yielded" extremely encouraging" results.
Chemistry's role is not restricted to detection. Chemistry also can help neutralize land mines once they are found.
Currently, most mines are destroyed by detonating them with the help of high explosives. The operation is hazardous, requiring people trained in handling explosives. Besides, detonating metallic mines creates more metallic fragments to further confuse demining operations.
Chemicals can be used to inactivate land mines with less splatter, says DOD scientist Patel. For example, amines such as diethylamine and diethylenetriamine and metal alkyls such as diethylzinc and triethylaluminum can initiate reactions that will burn the explosive charge. The chemicals are commercially available and inexpensive, says Patel, but they react only with specific explosives. "We need chemicals that we can apply to many types of explosives," he adds.
"The beauty of these chemicals is that they don't detonate the explosive," says Patel. And they work even if the mines are buried. Air is not even required to sustain the fire because the explosive itself has oxidizing power in NO2 groups, he adds.
Delivery of the neutralizing chemical to the explosive charge is central to the technology. DOD's humanitarian demining program has come up with two delivery systems. Both involve remote-control firing of a bullet through a "gun" into the mine. One version can be positioned just above the surface of the mine. The other version can be used to target mines from a distance.
Still another area in which chemistry can contribute is in rendering mines safe for handling. "Suppose you have a mine near a house or on a bridge," says Patel. Blasting or burning at such sites is undesirable. What's needed is to disable the mine. Then the mine can be attached to a rope and pulled to a safe distance for disposal.
Patel says polyurethane foam - the same material used in building insulation - can be used for this purpose. The foam permeates the exposed parts of the mine, "freezing" the triggering mechanism as it hardens. (The same principle is used to render letter bombs safe, for example.) Bright color can be added to the foam to mark the mine's location.
For children, farmers, and displaced families who must live with the carnage wrought by land mines, solutions can't come soon enough. As C&EN went to press, the Conference on Disarmament still had not decided to put land mines on its agenda. Sen. Leahy said in January that if no progress is made in Geneva by June, he will urge the Clinton Administration to reconsider and become an active participant in the Ottawa process.
By contrast, the Canadian process is well under way. In mid-February, representatives from 85 countries, including the U.S., met in Vienna to discuss a draft convention prepared by Austria. That will be followed by a second session in Vienna in the spring, a third in Brussels in the summer, and a fourth in Oslo in the fall. The Canadians have offered to host the treaty signing in December.
With the DOD programs for revolutionary technologies as well as short-term solutions driving the technological track, the demining tool box is becoming better equipped but is still far from being complete. Says Walden of Operation USA: "We [will] have the tool box when we have effective mine detectors of one kind or another that are affordable by a poor country and are being used by local people."
"I just wish we had more money" for R&D, said Leahy last December. Until effective ways to find and clear the mines are available, these vicious weapons will continue to be removed - as Leahy recalled a Cambodian saying to him - "an arm and a leg at a time."