Grantee Research Project Results
Final Report: Interferometric Continuous Emission Monitor for Active Control of Organic Emissions
EPA Contract Number: 68D99064Title: Interferometric Continuous Emission Monitor for Active Control of Organic Emissions
Investigators: Poulos, Arthur T.
Small Business: Optomechanical Enterprises
EPA Contact:
Phase: I
Project Period: September 1, 1999 through March 1, 2000
Project Amount: $69,853
RFA: Small Business Innovation Research (SBIR) - Phase I (1999) RFA Text | Recipients Lists
Research Category: Ecological Indicators/Assessment/Restoration , SBIR - Monitoring , Small Business Innovation Research (SBIR)
Description:
Active process control offers potential to reduce transient and steady-state emissions from combustors and incinerators (mobile and stationary sources). The goal is to alter combustion parameters (equivalence ratio, feed rate, diluent flow rate, combustion staging) in rapid response to the appearance of PICs and PIC precursors (Products of Incomplete Combustion). With an "intelligent" control system, pollutants may in principle be contained below maximum allowable limits. This objective is hampered, however, by lack of cost-effective "real-time" monitors capable of resolving key species concentrations from a complex exhaust mixture.Current methods of monitoring exhaust streams include non-dispersive infrared (for CO2, CO), chemiluminescent NOx detector, paramagnetic O2 analyzer, total hydrocarbon analyzers, gas chromatography, and near-IR Accousto-Optic sensors.
With the exception of gas chromatography, none of these techniques can measure organic products of incomplete combustion (PICs). While gas chromatography is extremely sensitive and selective to such species, it is not a real-time analytical device. Various researchers have studied use of tunable IR lasers and tunable infrared filters. However, these systems are quite expensive (lasers) and/or operate over only a limited spectral range, meaning only one or two species can be monitored at a time.
FT-IR spectroscopy appears to be the only general purpose technique having the spectral range and sensitivity required for real-time analysis of PICs. However, even field-portable FT-IR spectrometers suffer from sensitivity to vibrational misalignment, high cost, and relatively large size. Commercially available units are based on the Michelson interferometer design, which is inherently sensitive to vibration due to its separated mirror/beamsplitter optical arrangement. They are also inherently costly due to the difficult-to-fabricate KBr beamsplitter, and the use of a laser mirror tracking system.
The objective of this Phase I was to evaluate active control with a novel interferometer interfaced to a "trained" combustion control system. This particular interferometer uses a wavefront-division design and a step-scanning mode to achieve reduced sensitivity to vibration and background jitter, at relatively low cost and compact size. The research effort characterized the interferometric method with respect to its sensitivity, linearity of response, and accuracy for in-situ analysis of PIC precursors. In addition, methods were developed to integrate the interferometer into the feedback control system of a laboratory-scale two-stage combustor.
Summary/Accomplishments (Outputs/Outcomes):
The exit from the secondary zone was re-designed to permit optical access to unperturbed combustion products. Monitoring of the exhaust section was not considered appropriate because the combustion products are burned or otherwise modified in the afterburner. Also, sensitivity is reduced by dilution with excess air. In the new design, gases exiting the second zone are cooled by injection of small quantities of liquid nitrogen and mixing at a series of impinger baffles placed across the flow direction. The optical ports (window mounts) are located immediately downstream of the impinger baffles. They are followed by an afterburner and exhaust pipe to the hood system.
The spectrometer was first evaluated using quiescent calibration gas mixtures held in a 10 cm path length sample cell and placed on the mounting platform at the location corresponding to the combustion cross-flow. Three different types of samples were analyzed: acetylene-air, benzene- air, and a calibration mixture consisting of acetylene, methane, ethylene, ethane, and carbon dioxide. Limits of detection for acetylene and benzene were found to be 40 ppm and 4 ppm, respectively.
Also, several combustion gas mixtures were analyzed with the spectrometer assembly, which, in effect, simulated a real combustion run.
The interferometer data acquisition program was modified to provide a periodic reading of acetylene concentration from the measured spectrum. The combustion control program was likewise modified to accept readings from the interferometer program, and the two PCs were linked through RS-232 protocol and ms-comm commands. We successfully demonstrated error-free transfer of information between the interferometer and combustion control program, running in simulation mode.
Conclusions:
The proposed active control system consists of two interacting components: an in-situ FTIR monitor, and an intelligent feedback controller. The Phase I monitoring system was found to have sufficient sensitivity to detect products of combustion such as acetylene, ethylene, methane, carbon monoxide, and carbon dioxide, under fuel-rich conditions. The limit of detection of acetylene, an important soot precursor, was found to be 40 ppm. By comparison, homogeneous combustion of ethylene was found to produce acetylene at concentrations 78 ppm and 16,500 ppm at equivalence ratios 1.2 and 1.9, respectively. The limit of detection of benzene in the absence of interfering carbon dioxide was found to be 4 ppm, but detection at concentrations in real combustion streams will require higher spectral resolution and/or multi-component analysis. The feedback controller system was successfully adapted to accept concentration signals from the spectrometer, approximately every 50 seconds.The FT-IR spectrometer used in this project is capable of detecting major species such as carbon monoxide, water, and carbon monoxide under fuel-lean and fuel rich conditions. However, detection of hydrocarbon products of incomplete combustion is most realistic for fuel-rich conditions. Most industrial combustors and incinerators are operated close to unity equivalence ratio or even slightly lean. In a perfectly operating combustor, the PIC production would be lower than the sensitivity of an FT-IR based continuous emission monitor (CEM).
However, incomplete mixing and unpredictable changes in fuel or waste composition can result in high local equivalence ratios and attendant high concentrations of PICs. This phenomenon was studied by Brouwer, Sacchi, Longwell and Sarafim (Combustion and Flame, 99, 231 (1994)). In a reactor similar to that used in this project, Brouwer and co-workers observed that injection of small quantities of methyl chloride under lean conditions formed acetylene, ethylene and methane at concentrations as high as 1200, 2000, and 2900 ppm, respectively. Such concentrations are certainly in the range accessible to FT-IR detection at 10 cm path length. Indeed, PICs will generally be produced only when the combustion matrix contains fuel-rich pockets.
It is concluded that with further development, the wavefront-dividing interferometer can be engineered into a multi-species monitor for rapid feedback control of organic emissions.
Using a market-niche analysis from Foresight Science and Technology, a potential commercial application has been identified. It is recommended that the technology be implemented for combustion process control/environmental compliance.
There is a new generation of power plants that rely on better process control to optimize performance. Thus, there are both an economic and regulatory compliance driver facilitating adoption. Because OME has already developed the optical aspects of the technology and a flexible computer interface design under the auspices of NASA Phase I and Phase II SBIR programs, it would appear that rapid commercialization should be quite viable.
Although the EPA SBIR program Phase I dealt primarily with detection of benzene and acetylene (which may be measured in the 10-14 micron wavelength range), the device range may be extended to wavelength regions appropriate for detection of other combustion species. These include CO, NOx, CO2, HCl, NH3 and others, which are detectable in the near-mid IR, 3-5 micron range, as well as other hydrocarbon species in the "fingerprint" range, 8-14 micron. Added to the ruggedness and low cost of the unit, industrial and energy generation process control makes sense and is also compatible with the R&D focus of gas sensing-product/combustant identification.
Assuming equivalent performance, the end-user would be a combustion or process industry that currently uses relatively expensive traditional infrared equipment to analyze for incomplete combustion or impure feedstock gas streams. OME's technology will lower costs, expected as a function of higher efficiency, and lower initial capital investment costs. Improved ruggedness is a long-term advantage, as state of the art FTIR instrumentation having moving parts requires some maintenance and hence, enhanced labor costs.
Driving combustion use is a new generation of power plant technology. Gas turbines are one example. A recent survey by Diesel & Gas Turbine Worldwide indicated that power generation gas turbine order growth is unprecedented, with the order level increasing to 64,254 MW and 875 units, v 33,197 MW and 754 units in the previous year. The bulk of these machines are for combined-cycle and peaking power plants in the U.S.
Gas turbines are likely to find increased applications in the future. Furthermore, the Department of Energy has identified gas micro-turbine technology as priority technology, because it places generators at the usage site.
Supplemental Keywords:
continuous emission monitor, combustion feedback control, FT-IR spectrometer, exhaust gas detection, gas turbine power plants, micro-turbine generator., Economic, Social, & Behavioral Science Research Program, Scientific Discipline, Air, Waste, Ecosystem Protection/Environmental Exposure & Risk, Sustainable Industry/Business, air toxics, Municipal, Monitoring/Modeling, Technology for Sustainable Environment, Analytical Chemistry, mobile sources, Engineering, Chemistry, & Physics, Incineration/Combustion, Economics & Decision Making, Market mechanisms, monitoring, real time measurement, stationary sources, air pollutants, continuous measurement, products of incomplete combustion (PIC), benzene, air pollution, combustion technology, active process control, Products of Incomplete Combustion (PICs), organics, continuous emissions monitoring, combustion, measurement, combustion exhaust gases, Interferometric, coal combustion, power generation , real time monitoring, coal fired power plants, cost effective, incineration, organic emissions, measure, cost effectiveness, real-time monitoringThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.