Tracking Mercury as It Goes Up in Smoke
Along with visible smoke, coal-burning plants produce invisible pollutants, including elemental mercury and mercuric chloride. The Environmental Protection Agency (EPA) has adopted rules designed to reduce the amount of mercury released by coal-burning plants. Unfortunately, current monitoring methods require problematic sampling procedures.
Detecting gaseous mercuric chloride takes the right laser. Shown is a schematic of a frequency-converted fiber amplifier, with a microchip seed laser injecting light into a fiber amplifier. The output of the fiber amplifier is frequency-converted into the deep-UV to photofragment mercuric chloride and thereby detect the gas. FI = Faraday isolator; FC = fiber coupler; λ/2 = half-wave plate; BBO = β-barium borate crystal; KTP = potassium titanyl phosphate crystal; LBO = lithium triborate crystal. Courtesy of Alexandra A. Hoops, Sandia National Laboratories.
A better solution for detecting an invisible pollutant involves using laser light that directly probes the gas, according to researchers at Sandia National Laboratories in Livermore, Calif. A group there has shown that the emission from laser-created photofragments is proportional to the concentration of gaseous mercuric chloride.
The technique offers some significant advantages, said Alexandra A. Hoops, a senior member of the technical staff. “Optical methods are unique in their ability to probe the flue gas in situ, thereby eliminating the challenges and cost associated with sample extraction and conditioning.”
She added that an advantage of the technique is that investigators can tailor the laser wavelength and detection scheme to the molecule of interest. This flexibility means that the interference of other chemical species in the flue gas can be minimized.
EPA rules require 0.1 parts per billion sensitivity, and the current continuous monitoring of mercury emissions can achieve this, but the sensors require sample lines to transfer the gas to the analyzer. These lines can become plugged and rendered useless. Moreover, chemical reactions in the lines can alter the sample, obscuring the actual composition of the combustion gas.
There also is a potential problem with existing monitoring methods. Not all can distinguish one mercury species from another. Regulations do not require detection methods to differentiate, but speciation may become a necessary capability in the future because elemental mercury and mercuric chloride are not equal. “The different mercury species have different removal methods and different environmental and biological effects,” Hoops said.
To overcome these issues, the researchers developed a real-time noninvasive detection method. They exploited photofragment emission, which occurs when a sufficiently energetic photon is absorbed by and fragments a molecule. The fragments, in turn, emit at a characteristic wavelength. In the case of mercuric chloride, the absorption is tens of nanometers wide in the ultraviolet and is featureless, lessening the demand on the laser source.
In developing an instrument for field use, the investigators faced a problem. The laser had to be rugged, compact and lightweight while requiring only electricity — and as little of that as possible. Because of these needs, the group selected a frequency-converted fiber amplifier, seeding it with an Nd:YAG microchip laser from Poly-Scientific, now Moog Components Group of Blacksburg, Va., operating at a repetition rate of 6 kHz and a wavelength of 1064 nm. They ran the fiber output through nonlinear crystals, with the resulting beam emerging at 213 nm. Because this wavelength is not resonant with any mercury ground-state transitions, the emission at 253.7 nm that arises from the mercury photofragments provided a direct measure of the concentration of the parent mercuric chloride.
However, in choosing a fiber laser, the researchers ended up with output power lower than that of a conventional laboratory laser. Thus, each pulse generated less than a photon of photofragment emission, so the researchers used a photon-counting scheme with a Hamamatsu photomultiplier tube. An advantage of this approach is that the pulse height from the photomultiplier, which can vary, is not an error source. The technique does require counting enough photons to be sure of the statistics, which in turn means that the repetition rate must be fast enough to make the collection of those counts speedy.
As reported in the July 1 issue of Applied Optics, the researchers measured the photofragment emission in a cell with an environment similar to that of flue gas from a coal-fired power plant. They demonstrated the ability to detect mercuric chloride in the parts per billion range. They estimated that, with a 5-min signal collection time, they could hit the 0.1 ppb detection limit.
As for the future, Hoops reported that the laser and detector are being integrated into a cart-mounted instrument. Other efforts aim to improve the laser and extend the technique. “Work is also under way to develop a compact, narrow-linewidth laser suitable for detection of elemental mercury,” she said.
Contact: Alexandra A. Hoops, Sandia National Laboratories; e-mail: firstname.lastname@example.org.
MORE FROM PHOTONICS MEDIA