Grantee Research Project Results

Final Report: Compact, Low-Cost, Long Optical Path, Multiple Gas NDIR Sensor

EPA Contract Number: EPD04016
Title: Compact, Low-Cost, Long Optical Path, Multiple Gas NDIR Sensor
Investigators: McNeal, Mark P.
Small Business: Ion Optics Inc.
EPA Contact: Richards, April
Phase: I
Project Period: March 1, 2004 through August 31, 2004
RFA: Small Business Innovation Research (SBIR) - Phase I (2004) RFA Text |  Recipients Lists
Research Category: SBIR - Air Pollution , Small Business Innovation Research (SBIR) , Air Quality and Air Toxics

Description:

The goal of this research project was to develop and demonstrate a low-cost, compact, infrared spectroscopic method for determining the concentration and mass emissions of a volatile organic compound (VOC) constituent, and for approximating total VOC mass emissions. Nondispersive infrared (NDIR) spectroscopy, based on the molecular absorption of infrared light at specific frequencies, is a universally recognized analytical method for accurate compositional analysis of many vapor species. Ion Optics, Inc., demonstrated the feasibility of a low-cost NDIR solution for total and selective VOC measurements. A custom, optical resonant cavity was introduced to increase the path length of broadband source light (and signal) without compromising radiation throughput. The design affords a compact, highly fieldable, efficient optic housed within a robust package and suitable for low ppm VOC measurements.

Although hydrocarbon VOCs generally are characterized by relatively weak molecular absorptivities throughout the mid-infrared spectrum, the appeal of the infrared approach is its ability to couple into compound-specific absorption bands resulting in superior chemical selectivity compared to other methods. The measurement requires only the interaction of light with the effluent so that the sensor is not in contact with the vapor or subject to deterioration or poisoning. In addition, the infrared approach is capable of very fast response times. This technical approach addressed both the challenges of weak signal modulation in the presence of ppm levels of VOC gases, as well as the challenges associated with selective identification of a specific VOC constituent. Methane was used to simulate general hydrocarbon VOCs, and toluene was selected to represent a specific constituent.

Summary/Accomplishments (Outputs/Outcomes):

A custom, efficient, long-path optic was optimized, assembled, and used for the detection and measurement of both total and specific VOC constituents. In addition, a method for measuring gas flow rate and determining mass emissions using this technology was presented. To overcome the signal-to-noise challenge (i.e., the detection of low-ppm concentrations of a weak absorbing gas), an optical pathlength greater than 40 cm was necessary. In addition to providing a long optical path, the design also required use of a conventional (nonlaser) broadband, incoherent light source, and had to be small and compact. To meet these goals, Ion Optics employed a faceted, cylindrical, optical resonant cavity whereby divergent, broadband light is efficiently reflected around the cell, providing very long path lengths within a small volume. The optic was assembled with a commercial infrared light source and a customized multichannel detector and used for VOC measurements.

Gas testing was conducted by flowing diluted methane (to simulate a general VOC) through the small volume defined by the folded-path optical cell. Methane (nominal concentration 20%, balance N₂) was diluted to concentrations ranging between 1,000 ppm and 5 percent by controlled mixing with a separate high-purity N₂ source gas using precision mass flow controllers. The infrared light source was electrically pulsed at 5 Hz, yielding a minimum sensor response time of 200 msec. The sensor’s dedicated hydrocarbon channel, responsive to all VOC gases, exhibited signal step changes as methane was varied from 1,000, 2,000, 10,000, 20,000, and 50,000 ppm. From these data, the extrapolated sensitivity of the dedicated hydrocarbon line was approximately 50 ppm.

Next, the sensor’s sensitivity and selectivity to a specific VOC constituent was demonstrated using toluene. Diluted toluene vapors in nitrogen, from 4 percent down to 2,000 ppm, were passed through the sensor. High-purity nitrogen source gas was bubbled through a hermetically sealed vessel containing commercial-grade toluene to generate the ppm levels of toluene. Gas concentration exiting the bubbler was estimated from the equilibrium vapor pressure. The VOC sensor’s dedicated toluene line showed strong response to the introduction of toluene and then clear step changes associated with changes in concentration. The extrapolated sensitivity was approximately 1,000 ppm. The sensor also showed excellent selectivity, because no signal was apparent when toluene was purged and methane was sourced through the gas cell.

Finally, a method of estimating total (and specific) VOC mass emissions was illustrated. Specific VOC mass emissions could be estimated simply by knowing the compound, concentration, and flow rate of the target VOC constituent. In the case of total VOC mass emissions, there must be some prior knowledge about the major components and their relative concentrations, which may be derived from previously taken calibration spectra on the exhaust. In either case, if the sensor can measure flow rate, mass emissions may be estimated. To measure flow rate, a tracer gas and a secondary sensor located downstream from the VOC sensor were used. To determine flow rate, each sensor needs only to detect the initial presence of the tracer gas. Using the time between detection events and the volume separating the two sensors, flow rate can be estimated.

CO₂ was selected as the tracer gas because it is a strong absorber and allows use of a secondary, low-cost sensor positioned down stream from the VOC sensor. With the two sensors in position, a pulse of CO₂ was released through the system at a controlled flow rate of 100 sccm. Signals resulting from CO₂ were recorded with the VOC sensor using the dedicated CO₂ line and the secondary sensor. From the temporal response of the gas puff, flow rate was estimated at 95 sccm, well within the error of the experimental setup.

Conclusions:

Using a custom optical gas cell, strategically selected bandpass filters, and an inexpensive infrared source, a compact, highly fieldable NDIR VOC sensor was developed. The sensor exhibited sensitivities approaching 50 ppm, very good selectivity, very fast response times, and was used to illustrate a method for mass emissions measurements. The price of the current prototype with custom optics and components falls below the U.S. Environmental Protection Agency’s (EPA) threshold capital cost of $15,000. This cost could be reduced substantially by developing molded optics and producing the other components in quantity. The technology as is addresses EPA’s need for low-cost, automated monitors for total VOC in exhaust streams, and could be used to quantitatively assess VOC mass emissions for specific constituents. The technology also addresses the need for hazardous air pollutant residual risk fenceline monitors, because it could provide rapid identification and measurement of VOC emissions at a wide range of source locations, including individual pollution control devices and facility fencelines. Specific products of interest to EPA that would result from this research project if Phase II was funded include refinery valve leak detection sensors and gasoline storage and vapor recovery systems control sensors. Ion Optics has identified several other niche market opportunities that currently are being pursued.

Supplemental Keywords:

nondispersive infrared spectroscopy, NDIR, volatile organic compound, VOC, hazardous air pollutants, monitoring, fenceline monitor, air pollution, emissions, hydrocarbon, methane, toluene, long-path optic, gas, sensor, CO₂, SBIR,, RFA, Ecosystem Protection/Environmental Exposure & Risk, Scientific Discipline, Air, Environmental Chemistry, Engineering, Chemistry, & Physics, Analytical Chemistry, Monitoring/Modeling, Atmospheric Sciences, Environmental Monitoring, diesel exhaust, aerosol particles, air quality, chemical characteristics, gas chromatography, particle size, particle sampler, hydrocarbon, emissions, atmospheric chemistry, long optical path, chemical detection techniques, monitoring, air quality model, particulate matter mass, ambient emissions, aerosol analyzers, VOCs, diesel & gasoline emission sampling, VOC emission controls