Emissions of air pollutants from industrial sources are of major public concern in Germany. Remarkable efforts have been made to control and reduce there emissions. A strong reduction of these so-called classical air pollutants like e.g. SO₂ out of channeled industrial sources could be recorded throughout the last years. However, at some industrial sites there are still several measurement problems that cannot always be solved appropriately by conventional measurement systems. For example, screening of fugitive emissions or sudden releases out of leakages is difficult to monitor correctly by conventional point sensors. In these cases the open-path FTIR method can show up with several advantages because of its measurement principle. In this paper, different open-path measurements at industrial sites are presented, including measurements of sudden releases of ammonia and measurements at and above an olefin production plant.

An advanced monitoring station was set up in the center of Milan, Italy. It is made up of several instruments for the measurement of atmospheric pollutants, including a DOAS system which is able to provide information of the time evolution of several primary and secondary atmospheric pollutants. A radioactivity monitor provides information about the time evolution of Radon daughters, thereby providing information about the evolution of the boundary layer. It is shown that pollution by primary pollutants can be described through a very simple model based on Radon observation. Secondary pollutants, like ozone and nitrogen dioxide can also be described by the same model. Observation in strong advective condition and during stability periods show that the presence of large concentration of nitrogen dioxide is due to radicalic processes which are also responsible for the formation of formaldehyde. The role of nitrous acid in the formation of radicals is also discussed.

A field study was performed to evaluate the strengths and limitations of open-path FTIR for the measurement of volatile organic compounds within operating process units. The infrared measurement pathlengths of 100 to 300 meters traversed pipe alleys and pump rows within the units. Three different gas chromatographic-based measurements were made in parallel with the infrared measurements. This was done to provide a verification on the accuracy of the infrared measurements and to provide a cross comparison among the various gas chromatographic (GC) measurements. Measurements at plants found extremely low (sub-ppm) concentrations of organic compounds within the units tested. FTIR results were generally lower than those observed by the three GC methods. Open-path FTIR measurements appear to suffer from a number of limitations at this time including problems with background correction, high detection limits and the impact of meteorological conditions. Further refinements in the technology are required before it can provide high accuracy/precision measurements. At present, best applications might be for leak monitoring and alarming.

The open-path FTIR (op-FTIR) measurement method has several characteristics, which make the method attractive for monitoring air pollutants in urban areas. In this paper recent op-FTIR measurements at an urban site in Germany are presented. These measurements include the first direct intercomparisons between op-FTIR measurements and an official measurement system of the Environmental State Agency in North- Rhine-Westphalia. These intercomparisons revealed very good results.

A set of data collected over the time period November 1993 to October 1994 is discussed in this paper. Although they are not a continuous data set, the FT-IR spectra were collected on 127 days during the year for a total 4463 spectra. These have all been analyzed for N₂O, CH₄, CO, O₃, and C₂H₄. Ethylene was measured on only a few occasions during the year and was only once substantially above the detection limit. The goal of this work was to determine whether any of the first four gases named above could be used as surrogate standards for precision and accuracy measurements for the FT-IR technique. This work shows that the daily variability of CO and O₃ is to high for these gases to be used for QA/QC purposes, whereas N₂O can be used. Under certain circumstances CH₄ can also be used, but only with great care, and in the presence of a local, varying source, it should not be used.

A pulsed-molecular-beam Fabry-Perot cavity Fourier-transform microwave spectrometer developed at NIST has demonstrated sensitivities for many polar gas- phase molecular species in the low parts per million (ppm) to parts per billion (ppb) range. The highest sensitivity is obtained using neon or argon carrier gas but nitrogen or air can also be used, with some loss in sensitivity (up to 100 times) due to the less efficient rotational and vibrational cooling in the molecular beam with diatomic gases. The minimum detectable concentrations for several representative compounds are provided. These include acetaldehyde, acrolein, propionaldhyde, benzaldehyde, p- tolualdehyde, methanol, SO₂, propene, methyl t-butyl ether, ethyl t-butyl ether, and others. Considerable attention has been given to making the instrument versatile and user friendly. The instrument is computer controlled using standard GPIB interfaces and several graphical interfaces under the C^PLUPLU operating system.

The Wisconsin Department of Natural Resources and Department of Transportation (WDOT) are conducting a joint study to determine the effectiveness of applying optical sensing techniques to vehicular emission monitoring. Two field studies using Remote Sensing Technologies, Inc. remote sensing equipment was conducted in 1993 and 1994. This paper describes the data handling and data validation activities of these studies, including identification of data elements. Data handling was performed by the same people who conducted the 180,000 vehicle emissions tests. A contemporary commercial spreadsheet from Borland International, Inc. was used to import the raw data from the remote sensor. The data was reviewed with the spreadsheet then moved into a Borland relational database product. The relational database permitted structured queries against databases of vehicle inspection/maintenance (I/M) data from WDOT, National Insurance Crime Bureau, and EnviroTest. We determined effective cut points for vehicles of different ages which delineated high-polluting vehicles (gross emitters) from vehicles in compliance. The I/M data was also used to intercompare the remote sensing results with traditional testing results. Remote sensing test results were then compared for errors of commission and omission with respect to I/M test. Ultimately, this remote sensing database technique could serve as a means for identifying gross emitters who would be required to visit an I/M facility for an out-of-cycle emissions test.

Spectroscopic techniques in the infrared and ultraviolet spectral regions may efficiently meet increasing measurement challenges in real-time detection of vehicle emissions in urban air quality studies. The results of a study are presented in which a Fourier transform infrared (FT-IR) spectrometer was used to continuously monitor motor vehicle exhaust emissions. The FT-IR identified several exhaust components, including ethylene, acetylene, propylene, isobutylene, the hydrocarbon continuum, carbon monoxide, carbon dioxide, methane, nitric oxide, and nitrous oxide. The emission of each species was shown to vary with vehicle type and the operating speed of the motor. Measurement results from this study support at least two commonly observed characteristics of motor vehicle exhaust. First, ethylene, propylene, and acetylene are common exhaust components. Ethylene was observed to be the most abundant and stable of the non-methane hydrocarbon emissions during idling for the three vehicles considered in this study. Second, the emission of NO as a function of time remains high and fairly constant at high speeds, while the concentrations of CO, HC, and the non-methane hydrocarbons decrease sharply. The results of this study strongly suggest that the FT-IR can serve as a continuous, real-time monitor for measuring motor vehicle emissions. Because it can simultaneously detect multiple pollutants and can operate in an automated fashion, the FT-IR represents a cost-effective means of determining the effect of vehicle emissions on air quality.

The Wisconsin Departments of Transportation and Natural Resources evaluated the hydrocarbon (HC) detection capability of the Remote Vehicle Emissions Sensing (RVES) system, which employs remote sensing technology, and Wisconsin's I/M analyzers, which use BAR90 specifications. Both analyzers employ non-dispersive infrared (NDIR) technology. Other recent research has quantified HC measurement inaccuracies for vehicle emissions analyzers that use NDIR technology or have BAR90 specifications. This research shows that BAR90 analyzers undermeasure some water- soluble HCs and NDIR analyzers undermeasure olefinic and aromatic HCs. This evaluation was based on both field measurements and calculations that simulate these inaccuracies. These calculations give a measurement accuracy value, which estimates the fraction of the total HCs in a vehicle exhaust sample that each analyzer measures. Other calculations quantify the ozone forming potential of this measured fraction by considering the reactivity of measured HCs. Our field measurements and calculations show Wisconsin I/M analyzer HC measurements are on average 7 percent and 1 percent less than RVES, respectively. Calculations estimate that both analyzers measure at most 43 to 71 percent (an average 61 percent) of the total HCs in an emissions sample. Additional calculations estimate that the HCs measured by both analyzers have 49 to 71 percent (an average 62 percent) of the ozone forming potential of the total HCs in an emissions sample.

Although FTIR is rapidly being recognized as a more reliable analysis technique that requires significantly less calibration than conventional analyzers for measuring automotive exhaust emissions, there is still a great deal of concern over the ability of FTIR data to correlate with the techniques presently in use. In this paper we will describe the results of several tests designed to compare the FTIR results with these other techniques: flame ionization detector for hydrocarbons, non-dispersive infrared for CO and CO2 and chemiluminescence for NOx. Measurements were made in the laboratory using a conventional US-75 test facility and other tests were acquired in parallel with the conventional analyzers for direct comparisons. In all of these tests the measurements were performed on diluted exhaust samples from a CVS system, but we will discuss the use FTIR for on-dilute combustion samples. We will also present results for several non-regulated species such as methanol, formaldehyde, N2O and ammonia. We will report on the results of using FTIR to calculate the non-methane hydrocarbon value (NMHC) and how FTIR can be used to directly measure other components in vehicle emissions or combustion processes. The overall correlation between the FTIR analyzer and the conventional techniques was quite good. In the case of the IM-240 test there were only two tests out of 45 where there was disagreement on the pass/fail assignment.

FTIR-Emission-Spectroscopy detects the thermal radiation of hot exhaust gases, yielding all information about its compounds during one measurement. Apart from the interpretation of smoke stack measurements, FTIR-Emission-Spectroscopy as a remote sensing technique was further developed for analyzing layered plumes, especially aircraft exhausts in a program of the German Science Foundation (DFG) on the effect of air traffic on the environment. The measurements shall be used as input data for model calculations and to validate the extrapolated emission data at flight altitude. The evaluation of the spectra with respect to the gas composition contains a line-by-line calculation of the transmittances of several layers of the exhaust plume (temperature- and concentration-gradients) followed by the radiative transfer through the medium towards the detector. The spectral input data are taken from the HITRAN 92 database. After the spectroscopic determination of the plume temperature and its profile from the CO₂-band around 2400 cm^-1, one obtains the total mass of the single gas species in the field of view of the spectrometer. Comparing the measured data for CO₂ with the theoretical emission index from ideal stoichiometric combustion, one obtains the emission indices for the other measured species. Knowing the fuel consumption of the engine, one gets the emission rates of the compounds in g/s. Several engine types, old fashioned engines (no bypass) and modern JT8 and CFM56 bypass engines at different thrust levels have been analyzed. H₂O, CO₂, CO, and NO concentrations can be derived immediately from the measurements right behind the nozzle exits, where the temperature profile is known to be homogeneous. The retrieval of the measured data far behind the nozzle exit uses the multilayer plume model. Formaldehyde and other hydrocarbon species are seen in the spectra and shall be implemented in the computer code. Apart from future applications for the turbine development and the engine-status control after a certain flight time, this remote sensing system can deliver emission data of aircraft engines and the temperature decay of the exhaust plumes at all altitudes.

Optical remote sensing and iterative computed tomography (CT) can be combined to measure the spatial distribution of gaseous pollutant concentrations in a plane. We have conducted chamber experiments to test this combination of techniques using an Open Path Fourier Transform Infrared Spectrometer (OP-FTIR) and a standard algebraic reconstruction technique (ART). ART was found to converge to solutions that showed excellent agreement with the ray integral concentrations measured by the FTIR but were inconsistent with simultaneously gathered point sample concentration measurements. A new CT method was developed based on (a) the superposition of bivariate Gaussians to model the concentration distribution and (b) a simulated annealing minimization routine to find the parameters of the Gaussians that resulted in the best fit to the ray integral concentration data. This new method, named smooth basis function minimization (SBFM) generated reconstructions that agreed well, both qualitatively and quantitatively, with the concentration profiles generated from point sampling. We present one set of illustrative experimental data to compare the performance of ART and SBFM.

Characterizing indoor air quality requires instrumentation with certain capabilities. Most importantly, indoor air analyzers must be capable of measuring low concentrations of airborne pollutants in complex environments. This capability and many others important to indoor air monitoring applications have been realized in a transportable extractive FT-IR gas analyzer that has recently become commercially available. Feasibility studies have been performed at a variety of sites to evaluate the potential of this new cell-based analyzer for indoor air quality analysis. This paper describes the instrumentation and methodology used to perform these studies and then presents the results of the studies. Through these results, certain conclusions become apparent concerning the instrumental performance features that are required to produce truly useful indoor air quality information.

We consider the design of a system combining computed tomography and Fourier Transform Infrared Spectroscopy (CT/FTIR) to detect and map the concentration of multicontaminant gas plumes in ambient air over a 100 m square area. Several factors affecting the accuracy of the reconstructed map and the detection limits that can be achieved in the field are discussed. The estimated cost and capabilities of the system are compared with those of a more conventional gas monitoring system that might operate over a similar spatial extent. The paper includes a description of a proposed system that is designed to produce a map of multiple gaseous contaminants with a resolution of 12 m X 12 m in a time of approximately 10 minutes by sequentially measuring the contaminant concentrations along 48 intersecting beam paths and then reconstructing the map using a CT algorithm adapted to detect Gaussian plumes. The optical elements consist of an FTIR mounted on a steerable telescope platform, a second remote steerable mirror platform, and 32 fixed retro-reflectors.

Continuous particle and volatile organic compound (VOC) monitoring in residential homes were combined to measure indoor air quality. Fourier transform infrared spectroscopy (FTIR) was used to monitor the VOCs. There are several advantages to this technique, including quantifying components, identification of unexpected components, following levels of these components with time, and identifying sources of specific VOCs. In addition, the relationship of human activities to VOC levels was determined. In addition to the FTIR measurements, we have made measurements on particulate levels which are important to consider for indoor air quality.

This paper reports on numerical studies to evaluate the use of a series of scanning open path FTIR spectrometer measurements coupled with computed tomography to create 2D maps of chemical concentrations in air. When taking open-path measurements in the field using a scanning system, depending upon the scan rate and resolution, each measurement is taken at a different point in time. Therefore, as measurements are obtained, the concentrations of the pollutants in air are changing over time and space. The computed tomography mapping system must accurately create a series of maps from air concentration profiles that are in a state of flux. This remote sensing/computed tomography system was evaluated using a series of test maps that simulated the generation and dispersion of contaminant plumes over time. Contaminant generation rate, wind speed, and wind direction were varied to create concentration profiles that changed every fifteen seconds over several hours. A computer simulation program calculated the open-path measurements using these test maps for different interferometer scan times, and a series of reconstructed maps were obtained. The reconstructed maps were compared with original test maps and were evaluated both qualitatively and quantitatively using four measures of image quality. Results of this research provide guidance as to the range of acceptable interferometer scan times that can be used to map concentrations over time using different meteorological conditions, contamination generation rates, and number of contaminant sources.

Beam path average data from an open path Fourier transform infrared (OP-FTIR) spectrometer can be used to reconstruct 2D concentration maps of the gas and vapor contaminants in workplaces using computed tomographic (CT) techniques. However, a practical limitation arises because many source and detector units are required to produce a sufficient number of intersecting beam paths in order to reconstruct concentration maps. A monostatic OP-FTIR system which is capable of rapid beam movement can be used to eliminate this deficiency. Instead of many source and detector units, a number of the intersecting folded beam paths can be obtained using many flat mirrors and retro-reflectors. We conducted tests of several beam configurations generated for a single scanning FTIR system using 54 flat mirrors and 56 retro-reflectors mounted along the perimeter walls of a typical sized 24 foot by 21 foot test room. The virtual source CT configurations were tested using concentration maps created from tracer gas concentration distributions measured experimentally in a test chamber. Computer simulations of different beam configurations were used to determine the optimal beam geometry. We found that high concentration areas and the general concentration gradient pattern could be resolved from tomographic reconstructions calculated based on 102 folded beam paths. However, the reconstructions showed some effects from noise and peak-smearing artifacts. The noise level could be reduced and the quality of reconstruction maps were improved by using a spline interpolation method to correct for the influence of folded rays. We refer to this approach as a virtual source CT geometry.

This paper presents studies conducted in an indoor exposure chamber, on the detection and mapping of chemical concentrations using open-path FTIR (Fourier Transform Infrared) spectroscopy and computed tomography. A vertical flow indoor air chamber having roller tracks at the periphery was used. The source and detector of an open-path FTIR spectrometer were mounted on separate tables place on the tracks. Sulfur hexafluoride was injected from the floor of the chamber and the plume was scanned using the FTIR spectrometer. Different ray configurations were used to scan and map the plume on a 2D plane. To test the validity of the FTIR/computed tomography system for mapping gases, point samples from within the chamber were drawn remotely from ports and were analyzed using an electron capture detector. The concentrations indicated by the gas samples at different points were compared to the tomographically mapped concentrations. A comparison of mapped concentrations with point sampled concentrations were made with respect to position, and concentration of the chemical peak. A comparison of different algorithms for mapping chemical concentrations was performed.

An OPSIS differential optical absorption spectrometer (UV-DOAS) open path demonstration system was operated over a 7 week period at an urban air monitoring site in East Hartford, CT. The study purpose was primarily to compare acetaldehyde and formaldehyde from the OPSIS to extractive data gathered by use of coated cartridges under an EPA recognized method at the same site. A secondary goal was to compare benzene, toluene, NO₂, and O₃ OPSIS open path data to data gathered by methods currently approved or recognized by the EPA. This was the first field test for acetaldehyde, and the study showed that OPSIS was not able to detect ambient acetaldehyde concentrations. However, the system was able to measure ambient levels of the other 5 pollutant parameters. The correlation of OPSIS data was excellent for O₃ and NO₂ and less well correlated for formaldehyde, benzene and toluene. The paper describes the monitoring equipment operational characteristics, equipment deployment, and calibration, and data comparison results. A brief summary is given on suggested improvements and retesting requirements.

This paper describes a technique involving the use of two open-path Fourier-transform infrared (FTIR) spectrometers to isolate emissions in complex source environments such as oil refineries or chemical manufacturing facilities. Typically, open-path FTIR monitoring involves collection or generation of an upwind (background) spectrum (I₀) prior to the commencement of measurement activities, and the use of that spectrum to create absorbances for subsequent downwind (sample) spectra (I₀) values. In situations where the upwind single beam spectrum is relatively clean or unchanging in time with respect to target compounds and interferences, this method of handling I₀ is generally satisfactory. However, in complex source environments where either I₀ or I values can change rapidly, the simultaneous use of two open- path units can yield substantial improvements, especially with respect to target compound detection limits. Although the use of two FTIR units can greatly improve the problem of background variation for target compounds as well as for background atmospheric components and interferences, it introduces a number of other problems resulting in difficulties in the analysis of the resultant absorbance spectra. These problems are discussed, and some recently collected field data showing the application of this technique is presented.

During late summer of 1993 a field study took place in Baytown, Texas, where a comparison was made between the Opsis open-path system and conventional point sampling monitors. Simultaneous measurements of the three criteria pollutants sulfur dioxide, nitrogen dioxide, and ozone were made during a time period of a month, including data on wind speed, wind direction, and temperature. This study was made in an area with numerous major local sources of pollution. The degree of agreement between the open-path system and the point measurement system depended on the gas being measured. Ozone, which is well mixed, showed very good intercomparison whereas SO₂ showed poor intercomparison from time to time. By studying the discrepancies as function of the wind direction, interesting results were obtained.

Methods of estimating VOC emission rates from a point source are being field tested by the University of Kansas, in cooperation with Region VII of the U.S. EPA and Kansas State University. The methods use path-integrated VOC concentrations, meteorological data, and a form of the Gaussian dispersion equation. VOC concentrations were derived both from a whole-air canister sampling method, with subsequent GC analysis, and from open-path FTIR measurements; estimated emission rates produced from the two analytical methods were compared. Canister-derived concentrations provided higher mean estimation accuracies than did FTIR measurements for both 1, 1, 1-Trichloroethane and toluene; however, for a third data set consisting of all other compounds released, FTIR measurements provided higher values. Estimation accuracy also was evaluated as a function of atmospheric stability and downwind distance; accuracy generally increased and variability decreased as stability increased; accuracy was better at longer than at 50 meters.

An intercomparison between DOAS and gas-chromatographic techniques for the measurement of the concentration of benzene and toluene in the air has been carried out in Milan, Italy during the winter of 1994. Measurements of the natural radioactivity due to Radon have been carried out for a direct description of the mixing properties of the atmosphere. The temporal trend of primary pollutants, including hydrocarbons, follow, as expected, that of radioactivity. The results show that large discrepancies between the two methods have to be associated with the stratification of the atmosphere caused by ground based inversions. Scatter plots of GC against DOAS data show acceptable agreements, while the nature of observed offsets in DOAS system need further investigation. The results also show that the measurements are self consistent as they fit the physio-chemical evolution of atmospheric pollution at the site of sampling. In addition, data obtained from DOAS may be used to describe the evaluation of vertical mixing through the ratio toluene/benzene. Data on this new and important aspect are also presented.

Open-path Fourier Transform Infrared Spectroscopy (OP-FTIR) was used to conduct an air monitoring survey at a Fourier manufactured gas plant (FMGP) superfund site. This survey was performed in support and at the request of the Waste Management Division, U.S. EPA, Region VII. A three day study was conducted in which the OP- FTIR instrument was operated at several fenceline locations around the perimeter of the site in order to detect and quantify benzene emissions. Results of the survey indicated that benzene emissions correlate very closely with site excavation and soil processing activities. Varying concentrations of benzene were detected were and quantified in the field in near-real time. Laboratory analysis of the field data also indicated concentrations of ammonia and toluene at levels above the instrumental detection limit.

This paper describes an automated FT-IR open path monitoring system that bas been installed at Tinker Air Force Base to monitor volatile organic hydrocarbon (VOC) emissions from the Industrial Waste Treatment Plant. Coordinated FT-IR and gas sampling measurements were performed to provide a basis for the development of plume dispersion calculations to predict emission source strengths and fenceline concentrations. Methods developed to perform this analysis are described.

The advantages of measuring open-path Fourier transform infrared (OP/FT-IR) spectra at low resolution are discussed both from a theoretical and experimental viewpoint. In general, the optimum combination of selectivity and sensitivity is found when the resolution is approximately equal to the average full-width at half height (FWHH) of the analytical bands. The FWHH of many bands in the vapor-phase spectra of molecules of medium size, such as chlorinated organic solvents, is approximately 20 cm^MIN1, so that a resolution of 16 cm^MIN1 is often found to yield the most accurate analytical results. The low baseline noise level found when spectra are measured at low resolution can allow room temperature deuterated triglycine sulfate pyroelectric bolometers to be used instead of liquid nitrogen cooled mercury cadmium telluride photodetectors for OP/FT-IR measurements.

The combustion stoichiometry or air to fuel ratio is a critical operating parameter in many processes. The stoichiometry and thus the combustion efficiency dictates the plant economics as well as the environmental impact. In this work the technique of Fourier transform infrared spectroscopy has been evaluated as a nonintrusive probe of CO and CO₂ concentrations, average line of sight temperatures and temperature profiles on a 500 Kw oil-fired combustion test facility. Results have been compared with facility temperatures and concentrations calculated from an equilibrium chemistry model. Both FTIR emission and absorption configurations were evaluated thereby allowing conclusions as to the potential of these techniques for large scale facility diagnostics. CO₂ concentrations and profile temperatures were observed to agree well with the facility measurements and chemistry calculations. CO concentrations were observed to be in poor agreement with the calculations. This result is believed to arise from incomplete mixing in the test facility burner although other facility effects cannot be ruled out. The presence of CO in the spectra at an air to fuel ratio of 1.05 indicated that the FTIR methods can be used to diagnose burner operation.

In the interest of developing practical methodologies for remote passive FT-IR analysis of sulfur dioxide in heated smoke stack plumes, IR spectra have been collected in a number of relevant experiments. Field data includes passive remote FT-IR spectra collected at a coal-burning power plant for which plume conditions were characterized by in-stack continuous emission monitors (CEMs), spectra collected of a controlled plume from a model spectrometer, into which controlled levels of sulfur dioxide could be introduced while the spectrometer viewed sky backgrounds similar to those behind the actual power plant plume. An extensive spectral data set has also been collected in the laboratory under controlled target and background conditions using a heated cell. Typical spectra are presented and the potential for characterizing many of the important factors involved in remote passive FT-IR analyses through controlled- condition experiments such as these is discussed.

The development of Open Path FTIR for use in industrial setting has necessitated techniques for obtaining valid background spectra. Measurements over extended path lengths have resulted in greater problems with handling atmospheric moisture correction. Some industrial environments may even have almost continuous levels of chemical species present. Background correction techniques to address these problems employing a near-field source, a fixed reference, or a 'rolling' reference are presented and compared. Field experience obtained at several major plant sites over a year and a half of operation is reviewed.

The objective of this study was to evaluate instrument performance with the ultimate goal of developing guidelines for procedures that can be used by instrument manufacturers and end users for quality control and quality assurance. A glass calibration cell was placed in the IR beam path and various performance parameters were evaluated. The parameters examined were gas flow rate, spectral noise in the region of interest, interferogram height, detector linearity, polystyrene peak positions, instrument accuracy, and instrument precision.

The single-beam spectra recorded with two commercially available monostatic FT-IR open-path monitors exhibited nonzero signal intensities in the wave number regions where water vapor and CO₂ totally absorb. Each instrument yielded a measureable spectral response with the transmitting/receiving telescope blocked. This response is attributed to stray light within the FT-IR system. In some cases the stray light contributed up to 17% of the total return intensity measured along a 414-m path length. Although the baseline region of the single-beam spectra can be corrected for stray light, its effect on the spectral absorbance is not uniform over the range of absorbance values recorded in long-path measruements. Thus, the concentration of species measured along the path cannot be easily corrected, which has a deleterious effect on the accuracy of the spectral data.

The need continues to exist for faster analysis results and lower minimum limits (MDLs) to ensure that the protection of public health is not compromised during air contaminant releases associated with the cleanup of hazardous waste sites. This has led to the development of techniques to optimize and automate the analysis software routines employed in the field for use with open-path Fourier-transform infrared (FTIR) spectroscopy. This paper describes the development and implementations of one such optimization and automation approach based on experience at approxiamately 20 hazardous waste and industrial sites over the past four years.

The German legislation has a standard list of minimum requirements to approve new measurement techniques to measure gaseous components. The tests are carried out by an independent laboratory and the test report is evaluated and determined to be accepted or refused by a federal committee. Each gaseous component is tested separately both in the lab and during a three month field test where two indentical systems are installed in the same duct and the readings are compared with extractive reference methods. The final approval states the lowest approved measurement range and limitations to fulfill this measurement range. The Opsis DOAS system was tested by TUV in Colone during the years 1990 to 1994 and have so far been approved to measure NO, NO2, SO2, H2O, NH3, Phenol, Formaldehyde, and Mercury for CEM applications.

Progress on the developement of a nonlinear curve fitting computer algorithm for data reduction of optical remote sensing Fourier transform spectrometer (FTS) data is presented. This new algorithm is an adaptation of an existing algorithm employed at the Arnold Engineering Development Center for the analysis of infrared plume signature and optical gas diagnostic data on rocket and turbine engine exhaust. Because it is a nonlinear model, the algorithm can be used to determine parameters not readily determined by linear methods such as classical least squares. Unlike linear methods this procedure can simultaneously determine atmospheric gas concetrations, spectral resolution, spectral shift, and the background or (I^o(omega) spectrum. Additionally, species which possess spectra that are strongly masked by atmospheric absorption features such as BTX can also be incorporated into the procedure. The basic theory behind the algorithm is presented as well as test results on FTS data and synthetic data containing benzene and toluene spectral features.

As part of a program to develop quality assurance (QA) and quality control (QC) procedures for open-path Fourier transform infrared (FT-IR) monitoring, ambient gas concentrations were measured daily over an 11-month period. Some of these data will be used to describe the seasonal diurnal variation of atmospheric gases. To validate these data, a daily protocol was established to verify the long-term performance of the FT-IR system. This protocol includes measuring the electronic noise, the intensity of the return signal, and the magnitude of the baseline noise; examining the features and profiles of the single-beam spectra; and determining the repeatability of the position and full width at half height of selected absorption bands. Several atmospheric gases, such as methane, nitrous oxide, and cabon monoxide, are present in virtually all spectra taken. The use of the ambient concentration measurements of these species for QA/QC purposes was investigated. Of these gases, the ambient concentrations of nitrous oxide are the most stable, which makes it the most promising gas for use as a surrogate standard. In addition to the tests that were used to determine the stability of the instrument, procedures were used to evaluate the validity of the analysis method.

In Germany, the application of continuous monitoring systems in industrial plants started about 1970. Main basis uniform federal regulations were established to set up a suitability test procedure. Therefore many requirements were defined and refined from time to time to keep up with the dynamic development and progress in measuring technologies. Today automated optical systems have assumed an important role in emission as well as in process monitoring. The possibilities and limitations of their application in waste incinerators will be illustrated. Concrete examples are given for different IR and UV spectroscopy technologies (GFC, DOAS, FTIR).

Efforts to establish reliable reference spectra for as many of the Title III hazardous compounds as possible are underway. This work is being managed through the EPA's Office of Air Quality Planning and Standards, Emission Measurement Branch at Research Triangle Park, North Carolina. The goal of disseminating this information is complicated by the bulk of the data required to cover the 189 HAP's and their spectral interfering compounds at various temperatures. An investigation was performed into a methodology for archiving and disseminating the spectral data that would be viable for the foreseeable future. A methodology using CD-ROM and the information superhighway was adopted as a possible mechanism for archiving and dissemination.

This paper discussed studies performed in a 35 foot outdoor Teflon exposure chamber in Pittsboro, North Carolina. The purpose of this research was to evaluate open-path FTIR data and compare it to a reference method varying several parameters: the method used for quantification, the library used for quantification, and backgrounds representing varying environmental conditions. The reference method used was GC- FID. The chemical evaluated was toluene over the concentration range of 5-30 ppm. Two different quantification methods were used in this analysis: traditional peak area method and interactive subtraction. All quantification was done manually. Two different libraries were used for quantification: the Hanst library (137 ppm-m toluene) and the EPA library (249 ppm-m) toluene. Several concentration data files were analyzed with the original background and then with subsequent backgrounds obtained from that same day. The subsequent backgrounds contained varying amounts of water vapor. The effect on quantification was evaluated. For concentrations of 20 ppm and below there was no significant difference between the peak area and subtraction method. However, for concentrations greater than 20 ppm, the difference between the two methods increased up to 20%. The difference between the EPA and Hanst libraries was statistically insignificant below 25 ppm. At concentrations greater than 15 ppm, the two libraries deviated and the difference ranged from 5% to 10% with the EPA results being higher. Water vapor was found to drastically impact the quantification. The magnitude of the effect varied between quantification methods and libraries.

Optimal removal of ubiquitous water vapor spectral bands from open-path FTIR spectra is a well-known challenge. One potentially advantageous approach involves de- resolving water vapor spectra of high spectral resolution and carefully matching them to field spectra. This approach is compared with the more standard approach in which 'upwind' water vapor field reference spectra are used. Computer software implementations of these two approaches are described. A program is described that performs wavenumber scale calibration on single-beam background and measurement spectra using water vapor band frequencies prior to creating an absorbance spectrum. An iterative, interactive program is described for the de-resolution approach. It forms a specified concentration-path high-resolution water vapor spectrum, de-resolves it using Fourier spectral manipulation to a specified resolution and apodization, and allows small wavenumber shifts to be performed. The matching low resolution water vapor spectrum is subtracted from a field spectrum. A third program is described that facilitates subtraction of field water vapor spectra from field measurement spectra. Significant progress has been made in implementing the de-resolution approach, however, the field reference approach currently gives superior results, especially in the 3400-2700 cm-1 region. Further progress could be made by acquiring high- resolution water vapor spectra of higher ppm-m value and automating the procedure.

A description is provided of design considerations for a forthcoming cooperative field measurement study to evaluate optical remote sensing technology for potential use in directly measuring and/or inferring industrial air toxics emissions. The general layout of these experiements (tracer releases from simulated volume sources) also lends itself to development of dispersion data that can be used to test and improve upon existing volume source models. Accordingly, point sampling arrays will be deployed downwind of the tracer sources in addition to the remote sensing instruments. Sponsors of the cooperative field study include the Office of Air Quality Planning and Standards, Emissions Measurement Branch, and the Office of Research and Development, Air and Energy Engineering Research Laboratory of the US EPA, and the American Petroleum Institute. The field measurements are scheduled to take place between December 5 and December 23, 1994 at an open field site near Chapel Hill, North Carolina.

The Kentucky Department for Environmental Protection has been developing on-site monitoring capability for the measurement of air pollutants. The department has purchased a mobile laboratory equipped with a GC/MS for point monitoring and a long-path Fourier transform infrared (FT-IR) remote sensor unit for monitoring air pollutants at different locations in the State. Prior to deploying the FT-IR instrument in the field, the instrument has been evaluated for precision and accuracy with 15 certified gases (CO, NO, NH₃, COS, CS₂, SO₂, (CH₃)₂S, acetone, benzene, CH₃OH, CH₄, CCl₄, CCl₃H, C₂H₅OH, and H₂S) against the vendor provided calibration spectra by using a 15 cm quality control internal cell. Results of this study are presented. Some other studies include the cases of strong spectral overlaps and structured spectral features. Results of some short-term field study at Calvert City, Western Kentucky are also presented.

Present and upcoming regulations are creating increasing demands for instrumentation to monitor air quality and emissions from stationaiy and mobile sources. Remote sensing instruments using optical techniques are rapidly gaining popularity over existing point source monitors for most applications. Differential Optical Absorption Spectroscopy, (DOAS) based on optical absorption in the near ultraviolet and visible portion of the spectrum provides an attractive technique for simultaneous measurement of a number of important pollutants. In particular DOAS can measure and differentiate the so-called BETEX group of aromatic compounds, including benzene which cannot be readily measured with remote sensing FTIR techniques because of severe interferences from the ubiquitous CO2 and H2O. The traditional DOAS system1 uses a telescope to transmit a beam of broad band light through the atmosphere. The transmitted beam is then received by a similar telescope at the other end of a long path and disbursed by a spectrograph. Light paths of about 500 m are commonly used for most molecules at concentrations encountered in rural and urban conditions. The light captured by the receiver is brought by fibre optics to the analyzer, which consists of a grating spectrometer and a rapid scanning device. The purpose of the rapid scanning device is to overcome a problem that faces any remote sensing method, viz, the turbulence or scintillations of the air along the path being measured, which occurs on time scales of 0.1 to 1 second. The instrument must make its measurements in a time scale faster than this in order to prevent blurring of the spectrum. The traditional DOAS instrument overcomes this difficulty by using a rotating wheel, containing some 20 identical radial slits which sweep past the detector. Since the wheel rotates at a speed of 300 rpm, successive spectra are obtained at a frequency of about 100 Hz, which is much faster than the frequency of air turbulences. Thousands of scans are added in the computer over sufficient period of time to achieve get the desired sensitivity. The combined spectra is then analyzed to obtain the best fit to a combination of reference spectra of the gases believed to be present in the air being sampled

CETREL--Empresa de Protecao Ambiental, is an environmental engineering company, which is owned by the member companies in the Camacari Petrochemical Complex, the largest petrochemical complex in Brazil. CETREL operates a centralized waste treatment plant, treatment and disposal facilities, an incineration unit, groundwater monitoring and air quality monitoring networks. The air monitoring network was designed based on mathematical modeling, and the results showed that the monoitoring of hydrocarbons is important not just within the complex but also at the area surrounding the complex. There are presently no regulations for hydrocarbons in Brazil, however they are monitored due to concerns about health problems arising from human exposure. The network has eight multiparameter monitoring stations, located at the villages nearby, where hydrocarbons are sampled with Summa canisters and subsequently analyzed with a GC/MS, using a Cryogenic trap at the interface. The open-path FTIR is used to monitor at the individual plants and in the areas in between because it is more efficient and costs less than it would to attempt to achieve the same level of coverage using the canisters. Ten locations were selected based on mathematical modeling and knowledge of the likely emission sources. Since August 1993, there have been five different measurement campaigns.

The IR and ultraviolet emission spectra of hydrocarbons and toxicants was measured and analyzed as compared to theoretical data at room temperature. Based on this data we constructed an electro-optical gas detector for monitoring low concentrations of flammable paraffins, aromatics, and toxic hydrogen-sulfide. The optical method uses two wavelengths at several spectral bands: the signal and the reference which is sampled at a region where the hazardous gas does not absorb at all. Our apparatus is an innovative system that provides fast and reliable explosion detection at different LEL levels. As well, it can provide identification of low concentration of toxicants in the range of parts per million. The apparatus includes a fire detection option that can offer at the same time an automatic activation of fire suppression or neutralization system. It can detect paraffins in the range between 0.03 to 20 LEL per 1 meter by using the infrared spectral band, and aromatics and hydrogen sulfide in the range between 100 to 2000 ppm per 1 meter by using the ultraviolet spectral band. At both regions the accuracy is about 20%. This open-path, line-of-sight gas detector can monitor and transmit an alarm signal prior to occurrence of fire or explosion.

On May 30, 1991, the Office of Air Quality Planning Standards of the U.S. Environmental Protection Agency (EPA) published proposed rules regulating the total non-methane organic compounds (NMOC) emissions from municipal solid waste (MSW) landf ills1. Specifically the regulation requires emission controls on any MSW landfill emitting more than 150 Megagrams/year (Mg/yr) (167 tons per year). All landfills with design capacities less than 100,000 Mg (111,000 tons) are automatically exempted. If a landfill's planned capacity exceeds the 100,000 Mg threshold, the NMOC5 must be either estimated or empirically measured on a yearly basis until the total NMOC emissions exceed 150 Mg/yr at which time the installation of a control system is mandatory. The same criteria would also hold for a closed landfill except that the estimate/measurement step would only occur once. The proposed regulation outlined a three—tier approach to be used for this determination.

In this contribution I describe two applications of Fourier transform infrared spectroscopy to the monitoring of atmospheric compounds. First, I present an overview of IR solar absorption spectroscopy of the atmosphere: a bit of history, environmental relevance, and its use as a tool to determine the composition of the atmosphere. I then report FTIR solar spectroscopy measurements carried out at ground level at National Center for Atmospheric Reserach and on airplanes employing a spectrometer of 0.06 cm^-1 resolution. Sample atmospheric spectra and fitting examples are presented for key species relevant to stratospheric chemistry and global change: ozone (O₃) a chlorofluorocarbon (CF₂Cl₂), a greenhouse gas (N₂O), HCl, NO and HNO₃. Second, I briefly describe urban air pollution measurements at an intersection with heavy traffic in Tucson, Arizona. Two FTIR spectrometers of 1 cm^-1 resolution were employed to carry out long-path open-path measurements of the CO/CO₂ ratio and SF₆. Two FEAT and two LPUV instruments were employed for ancillary measurements of CO, CO₂, NO, CO/CO₂ ratios obtained by FEAT exhaust emission and FTIR ambient air measurements are reported. The two-week campaign was supported by the US DOT and EPA.