By J.R. Wilson
WASHINGTON — While most airline passengers in the United States may not yet have encountered or noticed the change, the Federal Aviation Administration (FAA) is spending nearly $100 million to install and maintain a new type of security device to check carry-on baggage for traces of chemical explosives.
FAA officials have purchased trace detectors from three manufacturers for installation at every U.S. airport that uses X- ray equipment to examine carry-on luggage for weapons or bombs.
While the FAA requires every carry-on bag to go through an X-ray, airport security personnel today only take about 10 percent to an adjacent trace chemical detector. Security officials choose these bags largely at random, although they also may pull a bag if they have reason to be suspicious.
Trace detection systems find extremely small amounts of explosives on such things as luggage, purses, briefcases, radios, and computers.
The trace detectors from Barringer Instruments Inc. in New Providence, N.J., and from Ion Track Instruments in Wilmington, Mass., use a process called ion mobility spectrometry. Meanwhile, the system from Thermedics Detection Inc. of Chelmsford, Mass., is based on high-speed gas chromatography with chemiluminescence detection.
For the Barringer Ionscan 400, operators collect samples either by wiping a surface with a cotton cloth, or by using a battery-operated hand-held "vacuum cleaner" that uses a filter card. The operator then places the sample onto a tray, which slides into an analyzer. The machine detects and identifies explosives automatically within five seconds.
Similarly, operators collect the Ion Track Itemiser sample by wiping a surface with a paper filter or by using a filtered vacuum. The operator drops the sample into an analyzer, which confirms and identifies explosives within five seconds. The Thermedics EGIS 3000 machine collects samples in a similar way, yet takes 18-seconds to detect and identify any explosives.
Ion mobility spectrometry, in general, involves drawing a continuous ambient air sample over a semi-permeable membrane, which filters out everything but particles from chemical explosives. The remaining particles ionize on weak plasma, the resulting ionized molecules separate according to charge, mass, and shape, as they drift through the analyzer. The machine measures amplified current from the detector as a function of time and generates a spectrum. A microprocessor then evaluates the resulting spectrum for target compounds and determines the concentration based on the peak height.
The Thermedics EGIS system combines gas chromatography with chemiluminescent detection to separate and identify compounds in a complex mixture. The machine heats sample material into a gaseous form, adds it to a carrier gas, and introduces it into the "separation column." Here, the mixture separates into its individual component compounds by precisely controlled temperature cycling. These components pass to the chemiluminescent detector, which responds only to nitro-based compounds of certain structures in the sample, thus detecting any nitrogen-based high explosive.
While false alarms have been a serious problem with many of the systems that help security personnel check baggage. Yet trace detectors rarely indicate the presence of explosives where actually there is none.
"The FAA statistics we`ve gotten show a total alarm rate across several million samples at 0.3 percent," says Ion Track marketing vice president Paul Eisenbraun. "It`s very difficult to establish a firm false alarm rate from that, because we do get real alarms that do not indicate an explosive in the bag. We may get an alarm from someone who takes nitroglycerin or from military personnel or miners or others who work with explosives. That kind of contamination can be transferred to a bag by touch and may remain for some time."
And that is the secret to trace detection, he adds: "We believe it would be very, very difficult" to place an explosive in a bag without getting a detectable trace on the bag itself. A recent test "indicated if the devices are used properly in the manner in which they are used at U.S. airports, the odds of detecting any actual explosive is very, very high," Eisenbraun says. The test was at the Defence Evaluation and Research Agency at Fort Halstead outside London. It looked at novice and expert bomb makers, and its detailed results were kept secret.
Officials of the contractors and the FAA are reluctant to talk about what the systems will detect. Yet officials of Barringer have publicly stated their unit can detect TNT, RDX, PETN, Semtex, nitrates, NG, HMX and other chemicals in concentrations as small as 50 picograms (billionths of a gram).
Trace units also can detect illegal drugs such as cocaine, heroin, methamphetamine, LSD, PCP, THC and MDMA. However, FAA regulations forbid the use of airline security equipment for drug detection. Yet each of the companies has sold units to the U.S. Customs Service and other law-enforcement agencies for that purpose, including for airport use in areas separate from passenger security checkpoints.
"We`ve averaged about $12 million a year for the past three years for trace units and by the end of this year will have some 600 in place," says Anthony Fainberg, director of civil aviation security policy and planning for the FAA. "Our goal, by the end of 2000, will be to have more than 95 percent of all airport embarkation points covered." The detectors generally cost between $40,000 to $60,000.
The Thermedics EGIS III and IV, both introduced earlier this year, will be delivered to Germany and the EGIS IV to the United Kingdom in 1999.
Experts at the FAA, Air Transport Association in Washington, and contractors all agree the best design approach would be to combine the various detectors into single-pass systems. These would unobtrusively check passengers and their carry-on luggage for any potential contraband, from guns and knives to explosives. But to be feasible, such systems could not slow down processing of the nation`s more than 1.6 million passengers a day on some 20,000 domestic and international flights by scheduled U.S. airlines.
"My wish would be for us eventually to have every passenger walk through a trace portal," Fainberg says. "Technology probably won`t allow that for a couple of years yet. And eventually it also would be nice to be able to trace every carry-on bag, as well. But it probably will be another five to ten years before we will have a combination X-ray and trace detector."
All of the companies producing the current systems are working on one or more of those advanced capabilities, including some passenger portals that officials have already field tested in Albuquerque, N.M., and Boston. A new test will be later this year at McGhee Tyson Airport in Knoxville, Tenn.
A variety of portal designs are under investigation. Designers at Ion Track, for example, are working with the gas dynamics lab at Pennsylvania State University on a portal incorporating their patented "human convection plume" process.
According to the basic concept, the human body is a heat engine that produces airflow of about 40 liters per second. The university-developed equipment can visualize that, and follow-on research indicates it could detect explosive traces. The first Ion Track portal using that approach is to be ready for preliminary testing this month. If successful, four or five more units will go into field testing by year`s end.
Thermedics officials claim their SecureScan portal can handle eight to 10 passengers per minute and detect explosives, drugs, and metal simultaneously.
"The portal has wands that brush against you as you walk through," says Thermedics president Jim Barbookles. "That creates the sample input. The wands are a light extruded metal that are spring-loaded and brush against you as you walk through. There`s a cartridge inside the unit that rotates after each use and automatically changes the filter. On the wand is a series of holes that collect the sample."
Barringer president Ken Wood says while portals have been in existence for more than two decades — largely at nuclear power plants and similar facilities — they are too large, too expensive, or ineffective.
"Particles are the problem in collection," Wood says. "One unit uses paddles you pass through, which may encounter some passenger resistance. Others use airflows, including ours, but you have to be careful because you don`t want to put someone through a wind tunnel. So collecting those particles quickly and in a non-invasive way is a challenge. We`re looking at a system that will only require about three seconds within the portal."
Barringer officials seek to enable their portals to extract trace elements related to explosives from the airflow into which each passenger will step and pause for about two seconds. To do this, they are licensing airflow dynamics and preconcentration technology from Sandia National Laboratories in Albuquerque, N.M.
"We believe we have a good concept on how to make this high throughput in a real working environment," Wood says. He expects to have a production portal ready for deployment by the first quarter of next year.
Other systems are in development to detect trace chemicals on documents such as boarding passes, or special tokens that passengers take at the gate. Versions of these systems also are undergoing limited testing at small airports such as Knoxville. FAA leaders may make a decision on their future implementation within the year.
One consideration is key at U.S. airports: designers must not burden screening machines with so many sensors that their use slows passenger boarding. "Right now we`re attacking it by not testing everybody," Fainberg says. "If we did use (trace detectors) for everybody, it would be a serious problem because it takes up to 30 seconds. The goal of the portal detectors is to have it complete a check in about six seconds, which is about the same as it takes to run a bag through the X-ray unit."
Speed of screening must be foremost in the minds of security systems designers — particularly as new technologies become available to screen passengers and baggage in finer detail than machines can today.
"Three or four years ago, the technology was finally developed to detect low vapor quantities, basically by detecting minute quantities of explosive adhering to dust particles. If it weren`t for the dust, we wouldn`t be able to do this," Fainberg says. "It isn`t computer speed that matters in the trace but small spectrometers and high-speed gas chromatography, both of which have come along in the past five years.
The same technological advances may apply to the walk-through portal. "The idea there is to look for a more effective collection mechanism, reduce the time required, and be less intrusive," Fainberg says. "A lot of my colleagues are pretty optimistic about it, although I want to wait for some results. What we have to do now is improve the operational impact in terms of cost and speed.
New enabling technologies and advanced microprocessing capability thus have had a major influence on the hardware side of the airport security equation. But officials acknowledge it will take more than that to stop terrorism effectively.
"It is sometimes overwhelming, because we never really get ahead of it," says Susan Rork, managing director of security for the Air Transport Association. "We don`t want to scare the public, but we have a system of deterrents that is notable in not having had any incidents."
Rork cites a 1995 case "when the plot to blow up 11 aircraft over the Pacific was uncovered (when a fire broke out in the bomber`s apartment in The Philippines). When you really think about the worldwide network of intelligence and law enforcement, it was a fluke that that plan was uncovered, but nonetheless it was. So my view is we`re doing a fine job of deterrence, but you can`t measure it. You just don`t know how many plots may be going on out there."
The FAA is looking at the Thermedics EGIS 3000 for use in airports.
