JEFFREY DANNEELS is developing methods to help utilities better assess and address their vulnerabilities.

In the six weeks that have passed since September 11, Americans have stopped taking many things for granted. Among other worries, we no longer assume that our airports are safe from hijackers or our mail from bioterrorists. Federal agencies are moving quickly to put new, stricter security measures in place. But when it comes to protecting water reservoirs, researchers at the Department of Energy's Sandia National Laboratory are already one step ahead. For the past few years they have worked on sophisticated means to identify vulnerabilities in the nation's water infrastructure, as well as technologies to detect in real time contamination of the water supply.

"We started exploring the possibility of working together [with the Environmental Protection Agency (EPA) and the American Water Works Association Research Foundation (AwwaRF)] to enhance the security of America's water infrastructuresupply, treatment and distributionwell before the September 11 attacks on the World Trade Center and the Pentagon," Sandia scientist Jeffrey Danneels says. His efforts, which have taken on a new urgency since the recent terrorist attacks, have focused on establishing a program to target problem areas on-site at utilities and train the personnel to minimize any risks. He held a workshop for AwwaRF and the American Water Works Association (AWWA) this past November and has more scheduled to begin this month.

Danneels modeled his program after one Sandia initially created to support the U.S. nuclear security mission and that has since been adapted to assess the threat of terrorist attacks on government buildings, air force bases, nuclear power plants, nuclear processing facilities, prisons and federal dams. The EPA is most interested in analyzing the water distribution systems that serve the country's 340 largest cities. Because many of these systems are more than 60 years old, they have different structures and therefore different security concerns. Danneels says utilities must follow three basic steps to evaluate their own individual vulnerabilities: They must assess how well their system detects a problem, how well it can delay the spread of the problem and how well it can ultimately respond to the problem.

Where at least the first step is concerneddetectionsome of Danneel's colleagues at Sandia may soon provide real help. Cliff Ho and Bob Hughes have created a novel real-time gas- and water-quality monitoring system, made up of a tiny sensor array and weatherproof casing. Whereas traditional monitoring involves collecting samples of water, gas or soil and sending them for laboratory analysis, which can cost from $100 to $1,000, the new device performs its testing in situ and sends its resultsvia a computer at a collection stationto an interactive Web site.

"The electronic sniffer is a unique monitor that can be put directly undergroundin groundwater or soils where the humidity reaches nearly 100 percentand detect toxic chemicals at the site without taking samples to the lab," Ho explains. "It has the capability of detecting in real time undesirable chemicals being pumped into the water supply accidentally or intentionally. It will be able to monitor sites containing toxic chemical spills, leaking underground storage tanks and chemical waste dumps, potentially saving millions of dollars a year in the process."


ELECTRONIC SNIFFER placed underground can detect volatile organic compounds in a water or gas supply in real time.

The sensor array contains a collection of different chemiresistors to detect a range of volatile organic compounds (VOCs). To make the chemiresistors, Hughes first mixed commercial polymers that had been dissolved in a solvent with conductive carbon particles. He then painted this inklike mixture onto electrodes in specially designed microfabricated circuits. If VOCs are present, the polymers absorb the compounds and swell, which in turn changes the electrical resistance in the circuit. The swelling and change in resistance correspond to the concentration of the VOC. Once the chemical is removed, the polymers shrink back to normal. "By using four different kinds of polymersone for each sensorwe think we can detect all solvents of interest," Hughes says.

Ho and other team members devised the weatherproof packaging for the chemiresistor chipwithout which the device could not be placed in water or underground. "The package is modular, like a watertight flashlight, and is fitted with O-rings," Ho explains. "It can be unscrewed, allowing for easy exchange of components." All told, the casing, constructed of stainless steel, measures a mere three centimeters in diameter. Chemical vapors pass through the casing to the chemiresistor array through a small window covered with a waterproof Gore-Tex membrane. When the device is placed in water, VOCs will partition across the membrane into the gas phase.

The scientists recently placed the sensor at Sandia's Chemical Waste Landfill to see how well the device works outside the lab. There it is suspended about 60 feet down a screened well and logs data every hour. This field test will last for several weeks or months, and others are planned at Edwards Air Force Base and the Nevada Test Site. From the experiments, Ho, Hughes and their colleagues hope to determine the sensor's life span, as well as its performance when temperature, pressure and humidity vary.

"Over the next few years I expect we will see this invention being applied to DOE sites that require monitoring, remediation and/or long-term stewardship of contaminated sites, which currently spend millions of dollars for off-site analysis of manual samples," Ho adds. "This device can also be applied to numerous commercial sites and applications, such as gas stations, which include more than two million underground storage tanks that require monitoring to satisfy the EPA requirements."

And the electronic sniffer may offer at least part of the solution toward safeguarding the national water infrastructure. "A low security level might mean hiring a security guard and installing some detection features around critical assets, and that won't cost a lot," Danneels says. "But to stop a fairly organized group from committing a terrorist act could be extremely expensive."