By Steve Karsjen
This was no ordinary bake-off. There were no apron-clad chefs standing around taste testing entries in the chocolate chip cookie contest. In this bake-off, as it was affectionately called by participants, the "recipe" being judged was for the best continuous emission monitor, or CEM, and the prize was the enviable status of being judged the company with the best the CEM industry had to offer.
No small distinction, this honor would not only carry with it bragging rights but an opportunity to impact the way mixed wastes are monitored during the incineration process. Mixed wastes are those containing both hazardous and radioactive constituents. Disposing of mixed waste is an enormous problem for government agencies like the Department of Energy.
"It's estimated the United States has in excess of 165,000 cubic meters of mixed waste," says Bill Haas, a senior scientist at Ames Laboratory and a technical consultant to DOE's Characterization, Monitoring, and Sensor Technology Crosscutting Program, the program that along with the Environmental Protection Agency organized the bake-off. "That translates into approximately 800,000 55-gallon barrels of waste needing to be destroyed."
Mixed waste can be anything from tubing to hoses to paper products. Its common denominator is that the waste is hazardous and radioactive, mostly low-level and mostly the byproduct of government research or weapons production. Currently, the only EPA-approved process for treatment of mixed waste is incineration. As the largest operator of mixed waste incinerators nationwide -- with two sites at Idaho National Engineering and Environmental Laboratory, one at the Savannah River Site and another at Oak Ridge National Laboratory -- it's easy to see why the DOE would place importance on finding a safe way to dispose of mixed waste.
Key to operating a successful mixed-waste incinerator is the ability to monitor the materials, or feed-stock, that are fed into the incinerator in order to ensure the heavy metal emissions coming out of the incinerator's smokestack are within limits set by the EPA. Incinerator operators can monitor the feedstock two ways: by taking samples and sending them to a laboratory for analysis, which is laborious and time consuming, or by performing periodic trial burns every three to five years to verify that the facility meets EPA emission standards.
During a trial burn, the incinerator is run under a range of conditions, and measurements are made based on EPA standards. "As long as you're able to operate with emissions that are less than what the regulation says, you're OK," says Haas. "However, three to five years is a long time between measurements."
The beauty of CEMs, he says, is their "continuous" operation. "In fact, if everything works out as planned and CEMs are a mandatory piece of hardware on every incinerator, incinerators will not be able to burn waste if the CEM is not working," notes Haas. "What this will provide is a much higher level of protection to the neighbors related to the air they're going to be breathing -- air that's not been assaulted, even temporarily, above emission limits."
To determine the state-of-the-art in CEM development, Haas and other partners on the DOE/EPA bake-off team conducted a test in September 1997 at the Rotary Kiln Incinerator Simulator at EPA's National Risk Management Research Laboratory in Research Triangle Park, N.C. Results of the test were expected to help the EPA establish standards for smokestack emissions of six toxic metals -- arsenic, beryllium, cadmium, chromium, lead and mercury -- under the draft EPA Maximum Achievable Control Technology (MACT) rules for hazardous waste combustors.
During the demonstration, seven multimetal CEMs at various stages of development were tested, and results were compared to EPA's reference method for metals emission measurements. "A number of characteristics were taken into consideration in the test, such as sensitivity, speed and data quality, but the most important performance issue was whether the CEMs could quantitatively measure all six metals," says Haas. In the test, the six metals were injected along with fly ash from a coal-fired utility boiler into the afterburner of the EPA's incinerator simulator, and the relative accuracy of each multimetal CEM was calculated.
According to Haas, the results of the tests were "disappointing." None of the CEMs tested met performance specifications for sensitivity and data quality in EPA's draft MACT rule for hazardous waste incinerators. Only one of the CEMs tested was able to measure all six metals at the concentrations tested, but the relative accuracy of that CEM varied between 35 and 100 percent, not 20 percent or less as required in the EPA performance specification.
So what did the test accomplish? "Our findings were significant because they helped prevent the EPA from putting in place incinerator emission standards as part of its MACT rules based on the abilities of some monitors that weren't up to standards," says Haas. "Ultimately, we're not there yet in terms of satisfactory CEMs, but through our trials we're moving the process along by providing data from which regulators can make reasonable judgements about standards."
The North Carolina test did not go unnoticed by David Baldwin, Ames Lab chemist and director of the Lab's Environmental and Protection Sciences Program. Baldwin had provided a prototype high-resolution optical spectrometer as part of an inductively coupled plasma-atomic emission system for use by one of the groups involved in the test. Following completion of the tests, Baldwin and Haas began talking about the need to develop an effective CEM specifically for mercury.
Why mercury? "Mercury is near the top of the list of elements that EPA wants much better knowledge and control of," says Haas. "Mercury has a much greater importance than the other five elements simply because it's more common, more easily volatilized and has many pathways to human exposure."
Baldwin adds, "Mercury is a huge problem throughout the DOE." Measuring mercury in the waste feed is costly and difficult, especially for heterogeneous waste. Representative sampling is nearly impossible, and mercury is notorious for not being evenly distributed in the waste. Add to that the toxicity of mercury, which increases the risk of personnel being exposed to hazardous chemicals, and you have a major challenge facing the DOE.
"DOE is laden with tons of mercury that was used as part of the process to produce uranium at Oak Ridge National Lab," Baldwin says. "In order to decontaminate and decommission places like that, you're going to have to be able to monitor mercury to minimize exposure and danger."
Fortunately, elemental analysis is an area that has been at the forefront of Baldwin's work at Ames Lab, going back several years to when he and his colleagues used an inductively coupled plasma-atomic emission spectrometer to conduct on-site uranium analysis of soil at DOE facilities like the Fernald Plant in Ohio. When that work dried up due to a DOE shift in emphasis from on-site characterization and treatment of contaminated soil to bulk removal and transport to disposal facilities, Baldwin turned his attention to developing systems to support a new DOE thrust, which was analyzing air emissions resulting from incineration of mixed wastes.
"In order to do this, we needed to be able to draw material from air and then be able to analyze it with a smaller ICP than the one-ton spectrometer we had used for soil characterization," says Baldwin, who set about the task of developing a spectrometer system based on acousto-optic tunable filter and echelle grating technologies that would be smaller and provide higher resolution than past spectrometers.
At the same time, Baldwin looked at combining this system with a portable air-plasma ICP-AES system, which was being developed at Mississippi State University, to provide the real-time sensitivity and resolution capabilities necessary to perform multi-element analysis in smokestacks.
"That's how we got to where we are -- from dirt to building a high-resolution spectrometer that can be placed on a smokestack along with an air plasma and can measure emissions from those stacks," says Baldwin. From there it was a natural progression to Baldwin's current effort to develop a mercury monitor. "Since mercury was an area in which we were doing exploratory work already, we decided to do experiments to see whether the instrument we built to monitor other metals could also be used as a mercury monitor," he says.
The mercury CEM begins with an ordinary pen lamp, which shows the many colors that result from excited mercury. The color, or line, of importance to the scientists is the 253 nanometer emission line because it is the one that's absorbed by mercury. "In order to measure the amount of mercury in a cell, you need to be able to measure how much absorption of light is occurring as a result of mercury vapor in the cell absorbing light," says Baldwin. The process sounds simple but it isn't because of two outside influences: interference and scattering.
For the mercury CEM to work effectively, it had to solve the interference and scattering problems. For example, Baldwin says when you burn hazardous materials in an incinerator, you often have molecular gases like sulfur dioxide (SO2) in the stream. This presents a problem for current CEMs because SO2 absorbs the same wavelength as mercury, which can throw off the measurement process. "If you're measuring mercury based on absorption of a single wavelength, you may also be seeing a lot of SO2, which you might think was mercury, but it isn't," says Baldwin.
The mercury CEM also overcomes the second problem, scattering. When chemical or mixed waste is being burned in an incinerator, particulate matter, such as fly ash, is naturally present. "Particles, of course, scatter light which creates another type of interference, this one being scattering rather than absorption as in the case of SO2," Baldwin says. "Essentially, we do a background correction for molecular interferences, such as SO2, and for scattering."
Baldwin says a third problem with conventional CEMs is that they're not "hardened" for field use. Typically, the light has to go through the incinerator smokestack and out the other side and into a detector. "It could be 100 degrees with 100 percent humidity in this environment, and the likely result will be that those adverse conditions will cause the monitors to clog or fail to calibrate properly," Baldwin says. "They just don't handle the harsh environment well."
Baldwin says what is needed is a high-resolution spectrometer that can measure multiple wavelengths simultaneously and is not prohibitive in size. The mercury CEM addresses both those issues by bringing a high-resolution echelle spectrometer into the equation. "With an echelle spectrometer, you can accomplish both tasks," he says.
Building on the collaborative efforts already existing between Baldwin and Haas, another Ames Lab scientist will soon become involved in helping test the mercury CEM. Glenn Norton, an Ames Lab associate chemist, has funding from DOE to look at commercial mercury analyzers for combustion gases.
"He has quite an extensive testing system set up for this type of analysis," says Baldwin, who plans to use Norton's testing facilities and test the mercury CEM being built in the laboratory. "We need to concentrate on making a system that will survive these hostile environments, and then build a fieldable system."
Baldwin's goal is to be ready the next time DOE and EPA conduct another bake-off. "Bill has mentioned there's going to be another mercury test next fiscal year," says Baldwin. "We hope to be there with a system that meets the EPA's performance requirements."
For more information:
David Baldwin, (515) 294-2069, dbaldwin@ameslab.gov
Bill Haas, (515) 294-4986, haas@ameslab.gov
Current research funded by:
DOE Office of Environmental Management
Last revision: 12/17/99 sd
Home | Comments
| Search | Disclaimer