BY HANK HOGAN
SAN DIEGO, Calif.—The use of chemical treatments, most notably chlorine, as well as filtration and other techniques clean up water and remove potentially harmful pathogens and other contaminants. But what no one knows for sure is if the water flowing to the tap at any given moment is free of harmful bacteria and other bad factors.
“Right now, I don't think there's a robust system that can do real-time monitoring,” says Kevin Morley, regulatory analyst in government affairs with the American Water Works Association (AWWA; www.awwa.org) —the largest organization of water supply professionals in the world.
That may be about to change. In early July, two companies announced they were teaming up to provide a real-time waterborne pathogen monitoring system. JMAR Technologies Inc. (www.jmar.com) is partnering with the LXT Group (Escondido, Calif.; www.lxtgroup.com) to create BioSentry, a system that will, among other things, monitor the water in utilities and in water bottling plants. Such testing could not only spot problems but also might allow for a more economical use of chlorine, thereby reducing cost and improving water taste.
Another anticipated use for BioSentry is determining quality assurance for high-purity water systems found in semiconductor, pharmaceutical and other cleanroom applications. Yet another contamination-control possibility includes monitoring the wash water in food processing as a check on the cleanliness of fruits and vegetables.
JMAR is providing production know-how and eventual product sales and support, while LXT supplies the technology and applications expertise. Plans call for beta units to be in the field by early next year with full production scheduled for later in the year. BioSentry has no price tag yet, but JMAR CEO Ronald Walrod says, “It's not going to be pricey.”
The real thing in real-time
Water production facilities now use several different techniques to ensure their product is free of contaminating pathogens. One technique is to sample the water as it flows by and send it to a lab, which then tests for the presence of any of a number of microorganisms. Such tests could involve culturing for organic contaminants. According to the AWWA's Morley, lab results typically come back the next day. Water producers also look for changing pH levels, and the only way to be sure what caused the change is to go through lab analysis—again, which takes a day.
Another technique, used particularly in high-purity water systems, involves liquid particle counting using a laser. A laser beam is sent through the flowing water, light is scattered by any particles within the fluid, and the light scattered at right angles is collected by optics, detected by a photodiode or other device, and then analyzed to determine the size and count of particles. One problem this technique is that it can only spot objects within a given size range, with no real information about the type of particle. The particle could be a speck of sand, which presents no health hazard, or it could be Cryptosporidium, a microorganism that sickened 400,000 people in Milwaukee in 1993.
BioSentry, however, claims to be able to do that determination in real time. Like existing liquid particle counters, the JMAR/LXT Group system is based on scattering laser light, but makes use of what is known as Mie scattering— predominantly in the direction of the incoming light and not dependent on the wavelength of that light. The white light from mist and fog is one example of Mie scattering. Horiba Instruments Inc. (Irvine, CA) and Malvern Instruments Ltd. in the United Kingdom make particle counters and other instruments that depend on Mie scattering, although these products are not specifically intended to be real-time waterborne pathogen monitors.
David Drake, senior partner and chief engineer at the LXT Group, says that what makes BioSentry possible are economical desktop computers with microprocessor speeds above 2-GHz, less expensive solid-state lasers, and the development of better algorithms. Drake says a laser that cost several thousands of dollars 20 years ago can now be bought for a few hundred dollars. Some of BioSentry's algorithms are proprietary to the LXT Group, and allow for the backward solution of the scattering problem and the reconstruction of more than just the size of the particle. Drake says the solution enables BioSentry to be more accurate in its assessment and produce few false positives—times that the system said a problem was present when nothing was actually wrong.
“You've got the shape and the size, the index of refraction, and internal structure; so, if you were looking strictly at size, you wouldn't do that much good—you'd end up with a lot of false positives, and you don't differentiate very much,” Drake explains. “But when you're able to look at all four parameters, you get a much more comprehensive picture.”
The BioSentry system also makes use of a solid-state laser at 660-nm wavelength for its light source, and utilizes a CCD detector to capture the scattered light. Detection is done at multiple angles, yielding the information needed to extract the four particle characteristics.
In operation, the monitor goes through a “learning phase.” Pure samples of a particular target, such as Cryptosporidium, are sent though and analyzed. BioSentry then operates in a classification mode. The flowing water is scanned, and detected particles are classified into various types. According to Drake, BioSentry can also spot new suspects and alert operators that there are contaminants for which the system has not been trained.
The large-volume test
Other companies employing competing technologies are trying to solve the same problem as BioSentry, but moreover, BioSentry has yet to actually be produced for and used in a real-time mode for large flow applications. The ability to handle detection in large volumes of water is particularly important for bioterrorism.
“It's got to be able to pick out one spore in 100 liters—that level of sensitivity is critical,” says AWWA's Morley. He adds that it does little good if a system can only reliably detect contaminants at several times the level needed to make people sick. The ideal would be a device that could be placed in a water system and produce a confirmatory analysis of biological contaminants in real time.
Morley says the technology being pushed by JMAR and LXT is promising but needs to be tested in actual systems running in real-world conditions. Then water producers could gain faith that there would be a low number of false positives and could trust any warning generated. For their part, JMAR.and LXT plan to deploy beta units to several utilities early in 2005 for testing and validation. The company will then seek regulatory approval.
Speaking of the current technology, Morley remarks, “Depending on false positive rates, you have a certain level of confidence in what the red light means. I don't think we have a whole lot of confidence in the red light right now.”