AN INTRODUCTION TO Open-Path FT-IR Atmospheric Monitoring

ES&T

AN I N T R O D U C T I O N

TO

O p e n - P a t h FT-IR

Atmospheric Monitoring m here is growing ψ concern around the world about toxic and poten tially toxic chemicals in the atmo sphere. In the United States this concern culminated in the passage of the 1990 Clean Air Act amend ments (CAAA), which are aimed at reducing toxic air emissions by up to 95% (1). The final version of Title III of the CAAA lists 189 toxic air pollutants for which emissions must be reduced, and this number may increase in the future. Title V of the CAAA outlines a permit pro gram that is designed to ensure compliance with all CAAA regula tions (2). As this legislation and similar ac tions worldwide show, the respon sibility for improved air quality and proof of compliance with new regu lations is being placed on both in dustry and regulatory agencies. As a result, better air-monitoring meth ods are in demand. Although large emission sources such as smokestacks can be moni tored at the origin, it is impractical to monitor fugitive emissions from in termittent sources such as valves and pipe joints at the source. It is these difficult-to-characterize sources that the open-path Fourier transform in frared (FT-IR) spectrometry method is well suited to detect and monitor. As a screening tool for detecting fugi tive emissions and a compliance tool for monitoring fence-line concentra tions, open-path FT-IR can aid indus try and regulatory agencies in reduc ing emissions. 224 A

Open-path FT-IR spectrometry has been under investigation since the 1970s [2-3), but in the past five to six years the growing concern for the environment has pushed the method to the forefront of environ mental analytical techniques [4-8). Based on FT-IR technology that has been in use in the laboratory for more than 25 years, open-path FT-IR spectrometry is a reliable and versatile means of monitoring mul-

Environ. Sci. Technol., Vol. 28, No. 5, 1994

Perhaps the greatest advantage of the open-path FT-IR m e t h o d is its versatility. T I M O T H Y L. M A R S H A L L C H A R L E S T. C H A F F I N AeroSurvey, Inc. Manhattan, KS 66502-0012

R O B E R T M. H A M M A K E R W I L L I A M G. F A T E L E Y Kansas State University Manhattan, KS 66506-3701

tiple volatile organic compounds (VOCs) simultaneously. Why open-path FT-IR? In this photo (p. 226A) of a com mercially available open-path FT-IR system, the interferometer and de tection system mounted on the col lection optics are in the foreground. In the background are the infrared source and source collimation op tics. The versatility of sampling ge ometries that can be achieved using the open-path FT-IR method are clearly shown. The source and de tection systems require only an un obstructed line of sight between them in order to monitor for VOCs that pass through the beam of infra red radiation. This allows the scien tist to extend or shorten the path to fit the monitoring site. Perhaps the greatest advantage of the open-path FT-IR method is the versatility that is inherent to infra red spectroscopy. Because infrared absorption spectroscopy is based on molecular vibrations, most of the chemical compounds of interest in atmospheric monitoring will dis play a unique infrared spectrum. As Figure 1 shows, the outstand ing features of the single-beam transmission spectrum are the wa ter and carbon dioxide absorption bands that absorb all of the infrared energy in certain regions of the spectrum. However, there are three approximate regions in the spec trum that are useful for atmospheric monitoring: 700-1300 cm -1 , 20002250 cm"1, and 2400-3000 cm"1. An estimated 135 of the 189 species

0013-936X/94/0927-224A$04.50/0 © 1994 American Chemical Society

A commercially available open-path FT-IR system in use at a soil remediation site

currently on the air toxics list of the C A A A d i s p l a y c h a r a c t e r i s t i c absorption h a n d s w i t h i n these regions. The open-path FT-IR m e t h o d allows t h e s i m u l t a n e o u s detection of these c o m p o u n d s b o t h qualitatively and quantitatively in all but the most c o m p l e x s a m p l e matrix. Figure 2 illustrates this capability. Figure 2 s h o w s an absorption spectrum determined from two single-beam spectra collected d o w n w i n d of refrigeration units. Absorbance spectra are determined as the negative of the logarithm of the ratio of the sample single-beam spectrum to the reference single-beam spect r u m . (Choosing a reference spect r u m will be discussed later.) The field spectrum in Figure 2 is overlaid with laboratory calibration

FIGURE 1

A single-beam transmission spectrum of the ambient atmosphere 3

a

226 A

Collected from 400-4000 cm - 1 at 0.5 cm - 1 resolution over a 100-m path with a commercially available open-path FT-IR system.


absorbance spectra of Freon 22 (chlorodifluoromethane) and trichloroethylene. Each compound displays a distinctive "fingerprint" spectrum. This case is a simple example of qualitative analysis using open-path FT-IR spectra. However, interference among bands of more than one species in a single spectrum can occur in more complex mixtures. Another advantage of the openpath FT-IR method is that it is a long-path technique. Point samplers or monitors, such as a Summa canister or portable gas chromatograph, collect a sample from a distinct point in space; the sample in the open-path FT-IR method is the three-dimensional space defined by the infrared source and the infrared detector. Therefore, a plume can pass through the beam path at any point and the components of the

plume—if concentrated enough— will be detected by the spectrometer. This increased sampling volume means that a single open-path FT-IR system, rather than numerous point samplers, can be used to locate and define a toxic gas plume. As a result, a site can be characterized using the FT-IR method even when little site information is available prior to the investigation. As a long-path technique, the open-path FT-IR method is well suited for monitoring the boundaries of an industrial or hazardous waste facility for fugitive emissions. The open-path FT-IR method can also measure in situ, which eliminates sample handling and thus minimizes sample contamination. This feature is an advantage over traditional point sampling techniques, which require the sample to be collected on site and transported

to the laboratory for analysis. Finally, the open-path FT-IR method measures in near real time. This feature allows the technology to be used in process control situations where site activities can be altered to maximize efficiency and reduce emissions based on data obtained during the activity. The technology can also be used in systems that signal when excessive concentrations of toxic species present an imminent hazard. Traditional point samplers, such as absorbent tubes, usually are unable to perform this task because often the samples from these devices must be analyzed in a laboratory. The limitations of open-path FT-IR air monitoring have two basic origins: physical limitations of the hardware employed to make the measurement and data analysis limitations.

FIGURE 2

(a) Field spectrum collected downwind of refrigeration units in a European chemical plant (b) Laboratory calibration spectrum of trichloroethylene (c) Laboratory calibration spectrum of Freon 22


227 A

Physical limitations Configurations. The two types of open-path FT-IR hardware configurations commonly used in air monitoring are shown in Figure 3. Each configuration works well; their advantages and disadvantages will not be discussed in detail here, but rather the limitations of open-path FT-IR technology in general will be emphasized. The photo of the system in use and Figure 3a show the configuration that is commonly referred to as "bistatic." This terminology is fitting because the source and source optics are separated from the collection optics and detector. The other configuration, shown in Figure 3b, is termed "monostatic" because the source, source optics, collection optics, and detector are together at one end of the infrared beam path, and the hardware at the opposite end of the infrared beam path consists of only a reflector optic. All components of both systems mount on tripods during monitoring, and the data collection and analysis are controlled by a computer system. The cost of a complete system, a p o t e n t i a l l i m i t a t i o n itself, is $60,000 to in excess of $100,000 depending on the configuration. In general, the monostatic systems cost approximately $25,000 more than the bistatic systems because of the additional optical components. Weather. An obvious concern when monitoring the environment is the effect of meteorological conditions on the monitoring system. For the open-path FT-IR method, the only weather condition that precludes data collection is fog. Fog water droplets scatter infrared radiation from the source, and thus the detector receives no usable energy. Significantly, rain does not preclude data collection as long as the optics in the system are made from the proper material. For transmission optics such as windows or beamsplitters, the best material is zinc selenide. It is rugged and nonhygroscopic. Potassium bromide optics, on the other hand, are extremely hygroscopic and can be permanently fogged from the adsorption of water outdoors. Overall, the quality of data collected during periods of rain may degrade because of temperature fluctuations affecting the instrumentation or the large amounts of water absorption observed in the spectra, but the data will still be usable. Temperature extremes are not as 228 A

FIGURE 3

The most common open-path FT-IR hardware configurations

large a problem as temperature fluctuations. An FT-IR spectrometer is very sensitive to thermal expansion and contraction. (Optical alignment within the spectrometer changes with thermal expansion and contraction of the materials used to construct the individual components of the spectrometer. This problem is magnified when the beam of infrared radiation does not cover the complete area of the infrared detector. This situation is encountered at times when using the open-path FT-IR method.) Therefore, the instrument should be allowed to warm up and reach a stable temperature before data are collected. However, data collected during temperature variations may still be usable; the quality may simply be degraded. Quality checks. Another obvious question when working outside the laboratory is how to determine whether the instrumentation is functioning properly on site. There are three basic procedures used to perform quality control/quality assurance checks on open-path FT-IR instruments. Measuring the absorbance spectrum of either a polymer film such as polystyrene or a known quantity of a calibration gas in a gas cell in the optical path of the system allows the instrument operator to determine the frequency precision and absorbance reproducibility of the system. If these two parameters fall within an acceptable range (usually ±2%), the system is functioning properly. The third quality control/quality assurance procedure is to simply measure the noise in the spectrum


at a predetermined frequency. Most problems in interferometric measurements manifest themselves in the spectra as noise. Therefore, if the noise of a system is below some benchmark, the system is most likely functioning properly. A common benchmark used is 0.001 absorbance units between 900 and 1000 cm"1. A basic hardware limitation common to both bistatic and monostatic configurations is the need for an unobstructed line of sight between the two ends of each system. This requirement can be dealt with easily at most sites, but geography should always be kept in mind when considering the o p e n - p a t h FT-IR method for monitoring. Location. Another limitation of both system configurations is size. Although open-path FT-IR systems are often described as "portable," "transportable" would be more accurate. The reflector optic for monostatic systems or the source and source optic for bistatic systems can be manually carried up to a few hundred meters. However, the detection systems of both configurations are too heavy to allow practical transport by hand for more than a few tens of meters. Therefore, a vehicle is needed to transport the systems between sampling locations. The system's power requirements also need to be considered. Systems that operate on either 12 V DC batteries or 120 V AC power or both are available in both monostatic and bistatic configurations. The 12 V DC systems are more portable and versatile, but the batteries are a potential source of sparks and cannot be

used where there is a high risk of explosion. Monitoring environments such as a solvent storage area is one application where the monostatic configuration has the advantage over the bistatic configuration. Because monostatic configurations require power at only one end of the system, the reflector optic, which has no power requirements, can be placed inside high-risk areas while the detection system remains outside. The bistatic system lacks this flexibility without adding optical components. A final limitation is the path length that can be monitored. This will depend on the site geometry and the optical components used to generate the infrared beam. Optical systems can be assembled for path lengths of more than a kilometer, but they will be expensive, large, and impractical. The practical limit of path length for most commercially available systems is approximately 500 m. Data analysis limitations The fundamental information obtained from open-path FT-IR measurements is the identity and concentration of a chemical species averaged over a given path length. There is one general limitation to both the qualitative and the quantitative analysis: interferences caused by the open atmosphere, which may be a c o m p l e x m i x t u r e . Water, carbon dioxide, and many other compounds found in the ambient atmosphere absorb infrared radiation. Likewise, compounds may be present on industrial or hazardous waste sites that, although not of interest, absorb infrared radiation in the region of interest to the openpath FT-IR system. The absorption bands displayed by all of these compounds that are not of interest may hinder or preclude the analysis of one or more compounds that are of interest. This problem is not common, but it is prudent to obtain as m u c h information as practical about a given site and its likely contaminants before monitoring is started. Quantitation. The quantitation process begins in the laboratory and is based on Beer's law. Beer's law states that the absorbance resulting from a given species at any given frequency is equal to the product of the absorption coefficient (a constant) of that species at that frequency, the path length of the radiation through the sampling volume, and the concentration of

that species. In the laboratory, a vacuum system can be used to accurately introduce a known concentration of a gaseous species into a fixed path length gas cell. Once this operation is done, the absorbance spectrum can be determined and the absorption coefficient for the peak frequency of that species calculated. In actuality, spectra for four or more different concentrations of the species of interest are collected, and a Beer's law plot of peak absorbance versus the product of the concentration and the path length is made. The slope of this line is then determined using linear regression. This slope is equivalent to the absorption coefficient for that species at the peak frequency. (For a more complete discussion of the calibration process, see Reference 9.) This procedure works exceptionally well for data that are analyzed manually by a spectroscopist. However, some multivariate statistical routines that are used in data analysis of complex mixtures like those found in openpath FT-IR measurements require multicomponent calibration gases to be used for the calibration procedure. (For a more complete discussion of multivariate statistical analysis for FT-IR, see References 10, 11.) Library. The collection of compounds for which the calibration process has been completed is termed the library. Near-real-time results can be obtained only for those compounds that are present in the library. To obtain a library, either follow the calibration process described above for every chemical species that is of interest or is a possible interference, or purchase a general commercial library in digitized form and load it directly into the data analysis software. This commercial library is tempting because it is obviously less time consuming. In fact, there are applications that cover such a wide variety of chemical species that this is the only practical method of obtaining a library in a reasonable amount of time. However, absorption coefficients and band shapes observed for any given species can vary across instruments, and the commercial library will have been collected on a different FT-IR spectrometer from the sample spectrum being analyzed. Thus, the overall band shape observed in the library spectrum can be slightly different from the sample spectrum being analyzed. This difference in band shapes can

cause errors in the quantitative results. The magnitude of the errors can vary from species to species and instrument to instrument, and, depending on the experimental objectives, may not be significant to the overall experimental results. Reference spectrum. Once a sample spectrum is collected in the field, the first step in performing the quantitative analysis of that spectrum is to choose a reference spectrum. In the laboratory, a reference spectrum is obtained by evacuating the sample cell, filling it to atmospheric pressure with nitrogen gas, and collecting a single-beam spectrum. This operation results in an ideal reference spectrum because all of the experimental parameters are held constant, and the sample gas is the only difference between the sample and the reference spectra. In open-path FT-IR monitoring, collecting a reference is not so straightforward. The sample cell in this case is that portion of the atmosphere that makes up the threedimensional volume between the infrared source and the infrared detector. It is not possible to evacuate this volume, but there are methods of obtaining "clean" reference spectra. One of the most common methods is to collect spectra upwind of the facility of interest. Using the upwind spectrum as a reference to produce an absorbance spectrum allows the characterization of air emissions from the site. The largest problem with this technique is that, when a spectrum from one sampling location is ratioed to a spectrum from a second sampling location, the noise in the resultant absorbance spectrum may be high. The sources of this noise are the realignment of the open-path FT-IR system and changes in the concentration of water and carbon dioxide in the atmosphere. This increase in noise will degrade the detection capabilities of the system. A second means of obtaining a clean reference spectrum is to generate one using computer programs. There exists a commercially available database that contains all of the spectral lines of species that are present in the ambient atmosphere. This database can be used to create a single-beam spectrum for a given set of atmospheric parameters. The problems with this method are, first, matching the intensity and overall shape of the single-beam spectrum, and second, matching the exact peak frequencies of the ambi-


229 A

ent molecules. Noise can be introduced into the spectrum, which can degrade the detection capabilities of the system. A practical method for obtaining a reference spectrum is to use a spectrum that was collected along the same beam path as the sample spectrum. This is possible when the compounds of interest are transient. In actual field work, it is not uncommon to monitor a position for hours and detect nothing of interest. When a plume is detected in the beam, it is practical to use a spectrum collected during the previous hours along the same beam path as a reference spectrum. The advantage of choosing a reference spectrum in this way is that the resultant absorbance spectrum may have a larger signal-to-noise ratio and hence an improved limit of detection for compounds of interest. The problem is that, technically, the absorbance spectrum measured represents only a fluctuation in concentration of the species of interest, from the time the reference spectrum was collected to the time the sample spectrum was collected. Practically, this fluctuation in concentration is likely very close to, if not equal to, the absolute concentration of the species of interest. There is also one experimental method for obtaining a reference spectrum that involves applying a computer differentiation technique to the single-beam sample spectrum and using the resulting processed single-beam spectrum as a reference spectrum (12). This new method is relatively untried, and more experience is needed to establish its applicability. Once the sample absorbance spectrum has been determined, it can be analyzed to obtain an actual concentration for any absorbing species present. Here, too, there are separate methods. Analysis. The first is an interactive spectral subtraction in which the scientist employs computer software to strip the calibration spectrum of the species of interest out of the sample spectrum. This procedure produces a subtraction factor that can be used in combination with (a) the known concentration and path length used to collect the calibration spectrum and (b) the known path length over which the sample spectrum was collected to obtain a path-averaged concentration of the species of interest. Unfortunately, this method is tedious, is very time consuming, and relies on 230 A

human subjectivity to determine the subtraction factor. The other method is to use an automated spectral interpretation program. These programs build the sample absorbance spectrum using a linear combination of the calibration library spectra. The programs then minimize the difference between the linear combination of library spectra and the actual sample spectrum using a statistical least squares analysis. The result is an analysis factor that can be used to calculate the actual path-averaged concentration for the species of interest, similar to the way that the subtraction factor is used in spectral subtraction. One concern with this type of analysis is that the analyst can be insulated from the actual spectra and, therefore, be unable to judge the quality of data. Also, if the species is not present in the library but present in the sample spectra, the results for all species can be in error. One way to minimize these two errors is to review the original sample spectrum and the residual spectrum that remains once the linear combination created by the programs is subtracted. The history of a particular site can help to minimize this latter problem. Once the concentration of a species is obtained using the open-path FT-IR method, correct interpretation of that data requires that two fundamental characteristics of the data be remembered: concentrations are both path-averaged and time-averaged.


Interpretation. Figure 4 illustrates path-averaged concentrations. Because the open-path FT-IR method is a long-path measurement, there is no spatial resolution of the sample volume provided by a single open-path FT-IR system. (See Reference 4 for a procedure using more than one open-path FT-IR system.) So, as Figure 4 indicates, a 10 part per billion (ppb) plume over 100 m of the path will result in the same spectra as a 100 ppb plume over 10 m of path or a 1 part per million (ppm) plume over 1 m of the path. Thus, a traditional point sampler provides more spatial resolution information than a single openpath FT-IR system. Because the point sampler samples only one point in space, if a contaminant is detected, the analyst knows the location of the plume. The open-path FT-IR method does not locate the plume as exactly. Therefore, to obtain the most realistic picture of the plume, the shortest path length practical for monitoring the area of interest should be used. This arrangement minimizes the amount of sample volume that does not contain the plume of interest and provides the most realistic view of the concentrations within the sample plume. Of course, there is a compromise: When the location of a plume is not known, a greater path length will afford more sample volume and increase the likelihood of discovering a plume. The concept of time-averaged concentrations is analogous to that of path-averaged concentrations.

FIGURE 4

Illustration of the lack of spatial characterization of plumes inherent to open-path FT-IR

Concentrations obtained from a single sample spectrum are average concentrations over the time used to collect the spectrum. For this reason, short collection times will produce results that better represent the actual concentrations. This is one limitation that the open-path FT-IR method shares with the traditional point samplers. Signal-to-noise ratio. An additional consideration when choosing a collection time is the signal-tonoise ratio. This ratio is the driving factor in determining operational parameters because it determines the detection capabilities of the system. An individual single-beam spectrum can be collected in about a second on most systems. However, the noise in an individual single-beam spectrum would be too great for useful detection capabilities. Therefore, numerous singlebeam spectra are collected and averaged to form a single sample spectrum. This average reduces the random noise in the spectrum, and thus improves the detection capabilities. A guideline for choosing a collection time is to use the shortest time that results in a noise level that allows the detection capabilities required for the site of interest. One parameter that affects both the signal-to-noise ratio and the collection time is the operating resolution. In essence, the resolution determines the amount of detailed information that can be obtained from a spectrum. High-resolution spectra require longer collection times to produce a high signal-tonoise ratio, but provide more information about sharp absorption lines contained within the absorbance spectrum. Low-resolution spectra require a shorter collection time to produce a high signal-to-noise ratio, but only provide information on broad absorption bands. Therefore, the choice of what operating resolution to use in any given case depends on the objectives of the monitoring. (For a more complete discussion of operating resolution, see Reference 9.) It is important to remember that the resolution required to obtain the information sought from any given application will depend on the monitoring situation and experimental objectives. As an example, a typical monitoring situation will require the coaddition of 64 or 128 scans over a 90 s or 180 s collection time, respectively, at 0.5 cm"1 resolution. These spectra are stored on magnetic media at the end of each coaddition pe-

riod. The spectra can then be analyzed in the field to give near-realtime results, saved for data analysis at a later time, or both.

TABLE 1

Applications Numerous applications for the open-path FT-IR method have been developed. Transportable systems have been used to monitor hazardous waste and municipal landfills, industrial and municipal wastewater treatment plants, indoor air quality at facilities such as dry cleaners and print shops, industrial air emissions at oil refineries and chemical plants, fugitive emissions from active soil remediation sites, and many more sources of fugitive emissions. Further, open-path FT-IR systems have been developed for continuous monitoring of fugitive emissions at the boundary of industrial or hazardous waste sites. New applications of the openpath FT-IR method are a result of the versatile sampling geometries and long-path measurements achievable with the technique. However, these capabilities should not be the sole factors used in determining whether the technique is suitable for a given application. Rather, the detection capabilities of the system should be the determining factor. If an open-path FT-IR system cannot detect a certain species of interest at the levels of concern at a given site, it should not be used to monitor that site. A good predictor of whether the open-path FT-IR method will be able to detect a compound at levels low enough to be useful on a particular site is the detection limit of the system for that compound. The detection limit is the lowest possible concentration that will result in an absorbance spectrum that has a peak absorbance value for the compound of interest at least three times the peak-to-peak noise level in the baseline. The reason that the detection limits serve only as predictors is that the noise level and the interferences in the actual spectrum collected will determine the detection limit of the system for that situation. Typical detection limits can be calculated for compounds using the Beer's law plots described earlier and average noise levels found in field spectra. Some typical detection limits for compounds commonly monitored can be found in Table 1. Again, these values are only predictors. The actual detection limit of a system for a given monitoring situation cannot be de-

Typical detection limits for some common volatile organic compounds monitored by open-path FT-IR Compound

Detection limit, ppb "

Acetone Benzene Butyl methacrylate Carbon tetrachloride Chlorodifluoromethane Cyclohexane 1,2-Dichloroethane Toluene Triethylamine Vinyl chloride meta-, ortho-, para -Xylene

48 40 15 5 9 6 37 35 20 44 37, 17, 34

"The detection limits are calculated for spectra collected over 3 min at 0.5 cm - 1 resolution assuming the compound is uniformly distributed over a 100-m path.

termined until the monitoring has been performed. Conclusions Like any other analytical technique, the open-path FT-IR method has advantages and limitations. If used within its limitations, the method has an excellent opportunity to succeed and become widely accepted in the analysis of atmospheric pollutants. The extent to which the promise of the open-path FT-IR method will be realized depends on the skill and dedication of the scientists employing the technique as well as on the instrumentation itself. References (1)

" F a c t S h e e t s on Clean Air Act Amendments of 1990"; United States Environmental Protection Agency. Office of Air and Radiation: Washington, DC. (2) Hanst, P. L. Adv. Environ, Sci. Technol. 1971, 2, 91-213. (3) Herget, W. F.; Brasher, J. D. Appl. Opt. 1979, 18, 3404-20. (4) Hammaker, R. M. et al. Appl. Spectrosc. 1993, 47, 1471-75. (5) Kricks, R. J. et al. Proceedings of the 84th Annual Meeting of the Air and Waste Management Association; Air and Waste Management Association: Pittsburgh, PA, 1991; Vol. 2, Paper No. 91/57.11. (6) Green, M.; Seiber, J. N.; Bierman, H. W. Proceedings of the International Conference on Monitoring for Toxic Chemicals and Biomarkers; SPIE, 1992; pp. 157-64. (7) Russwurm, G. M. et al. "Use of a Fourier Transform Spectrometer as a Remote Sensor at Superfund Sites"; U.S. Environmental Protection Agency. National Technical Information Service: Springfield, VA, 1991; EPA/600/ D-91/115.


231 A

CALL FOR ABSTRACTS

*m ^ gjf u a n i j On-Site Bioreclamation The Third International Symposium April 24-27,1995 San Diego, California Sessions will cover new developments, ongoing research applications, and case histories of various in situ and on-site bioreclamation technologies and experiences. Interested parties should submit a one-page abstract to Rob Hinchee, Battelle, 505 King Avenue, Columbus, OH 43201, or FAX 614-424-3667. Abstracts will be peer-reviewed and papers will be selected for platform or poster presentation based upon technical merit and relevance to the symposium theme. Notice of acceptance will be sent after September 1. Call (US and Canada) 800-783-6338 or 614-424-5461 for more information.

Deadline for Submission: July 1,1994 fâSeâ

Battelle

U.S. Department of Εnergy Office of Environmental Restoration and W a i t · Management OH ice of Technology Development

Putting Technology To Work OHM Himidiatinii Services Corp.

CROUNWftTEH TECHNOLOGY.

Environmental PROTECTION

The Japan Research Institute, Limited

REFER TO KEY NO. 1

COUNT THEmYS

Do you k n o w h o w many ways there are to fulfill your giving wishes—and at savings that may surprise you? As you make plans that will influence your tax picture next April 15, request a copy of our helpful booklet, "Your Guide to Effective Giving in 1994." American Chemical Society Development Office 1155 16th Street, NW Washington, DC 20036 Q Please send me a free copy of "Your Guide to Effective Giving in 1994." Name Address City State

232 A

Zip


(8)

Yost, M. G. et al. Am. Ind. Hyg. Assoc. /. 1992, 53, 611-16. (9) Spartz, M. L. et al. Am. Environ. Lab. 1989, 11, 15-30. (10) Hong-Kui, X.; Levine, S. P.; D'Arcy, J. B. Anal. Chem. 1989, 61, 2708-14. (11) Saarinen, P.; K a u p p i n e n , J. Appl. Spectrosc. 1991, 45, 953-63. (12) Xiao, H.; Levine, S. P. Anal. Chem. 1993, 65, 2262-69.

Timothy L. Marshall (1) is the 1992-93 Phillips Fellow at Kansas State Univer sity and will receive his Ph.D. from Kan sas State University in 1994. He is the vice-president of AeroSurvey, Inc., and his research interests involve the devel opment of analytical methods and in strumentation for the analysis of com plex environmental and industrial hygiene samples. Charles T. Chaffin (r) will receive his Ph.D. from Kansas State University (1994). He is the president of AeroSur vey, Inc., a private company providing open-path FT-IR data. His research inter ests center around passive remote sensing using FT-IR spectrometry and extractive FT-IR measurements for environmental and industrial hygiene concerns.

Robert M. Hammaker (/) is professor of chemistry at Kansas State University. He holds a Ph.D. from Northwestern University and has served as a primary researcher at Kansas State for 30 years. His research interests are in molecular spectroscopy using Hadamard trans form spectrometry, Raman spectrome try, and open-path FT-IR spectrometry. William G. Fateley (r) is a university distinguished professor of chemistry at Kansas State University. He holds a Ph.D. from Kansas State and has served as editor-in-chief of the Journal of Ap plied Spectroscopy for 20 years. His research interests range from the devel opment of novel analytical instrumenta tion to the measurement of volatile or ganic compounds in the atmosphere using open-path FT-IR spectrometry.

AN INTRODUCTION TO Open-Path FT-IR Atmospheric Monitoring

Recommend Documents