High Temperature Reference Spectra
Donald G. Gardner
Lori T. Lay
The work reported herein describes efforts conducted at the Air Force's Arnold Engineering Development Center (AEDC) to establish high temperature reference spectra for some of the low vapor pressure compounds identified in Title III of the Clean Air Act (CCA) of 1990. To date, a high temperature infrared absorption cell has been designed and fabricated for this purpose. The cell has been incorporated into a high resolution FTIR spectrometer system and is currently being used to acquire absorption spectra of selected compounds at elevated temperatures. The measurement system provides spectral coverage from 500-4,000 wavenumbers (cm-1) in the temperature range of 70- 500 F. Details of the sample cell and preliminary spectral data of selected compounds are presented in this report.
A program to establish test methods for the 189 Hazardous Air Pollutants (HAPs) identified in the Clean Air Act (CAA) of 1990 has been undertaken by the U.S. Environmental Protection Agency (EPA). Fourier Transform Infrared (FTIR) spectroscopy is a technique with the capability to detect many of these compounds. Using suitable reference spectra this technique may be used to determine path integrated concentrations of many species. Currently EPA's Office of Air Quality Planning and Standards is using FTIR to collect data for Maximum Achievable Control Technology (MACT) standard development for many of the Title III air toxic compounds.
In the event that FTIR proves an acceptable technique for the detection of some of these compounds, reference spectra will be required for reduction of the spectra to path integrated concentrations. Most of the published reference spectra were obtained at ambient temperature and are suitable for open path monitoring applications. However, the use of ambient temperature reference spectra for the reduction of higher temperature gases spectra can lead to errors in the determined path integrated concentration. This error increases as the difference in temperature increases between the gas under analysis and the temperature at which the reference spectra were obtained. The temperature dependency is non-linear and is not the same for each specie. For open path monitoring this is not a significant problem due to the relative small temperature range observed in the atmosphere. However, for analysis of stack gases at industrial facilities which require a high temperature measurement such as in an extractive cell, this can result in significant error.
The use of reference spectra will eliminate the need for each laboratory to maintain a set of gas standards and gas handling equipment required for calibration of their FTIR measurement systems. 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. To establish the reference spectra requires the use of suitable ppm- level calibration gas standards. Such standards are easily prepared for many of the compound of interest. Compounds with vapor pressures sufficiently high to prepare quantitative gas standards are available from many of the specialty gas vendors. These standards are generally prepared and stored in high pressure gas cylinders. However, for many of the lower vapor pressure compounds no such standards are available. Special precautions must be taken when working with these low vapor pressure compounds. This involves preparation and analysis of the gas standard at elevated temperatures with the gas handling apparatus coupled to the spectrometer system. It is the apparatus and techniques for preparation and handling of the low vapor pressure compounds that are discussed in this report. In addition, preliminary absorption spectra of selected low vapor pressure compounds are presented.
AEDC is the world's largest facility complex designed to provide engineering development support for aerospace systems. Testing is performed on propulsion, aerodynamic, reentry, transatmospheric and space flight systems in an environment that simulates operational conditions. Understanding the test environment is essential to the AEDC mission. For many years engineers at AEDC have been interested in high temperature infrared spectroscopy of gases for the purpose of aerospace system diagnostics. Non-intrusive optical diagnostics of the test environment and propulsion system exhaust products requires an accurate knowledge of fundamental molecular parameters of gases such as H2O, CO2, CO and NO. Experimental determination of these parameters requires the use of sample cells that will function at high temperature conditions approaching that of the test environment. The high temperature infrared absorption cell used for molecular band model studies at AEDC have been operated over the temperature range of 300-1,000 K. With minor modifications this sample cell design is being used to develop high temperature reference spectra for some of the low vapor pressure compounds of interest to the EPA.
A BOMEM Model DA8.002 FTIR Spectrometer is currently being used in this study. This spectrometer system is equipped with a high performance vector processor system and a remote host computer with all necessary peripherals for data reduction. The system provides spectral resolution up to 0.004 cm-1. It is equipped with an output beam selection which enables the use of sample cells either internal or external to the device. The modulated output beam from the interferometer is directed to a focus inside the sample cell by means of the concave and flat mirrors in box 1 of Fig. 1. The diverging beam exiting from the sample cell is once again collimated by the optics of box 2 and directed into box 3. The off axis reflector of box 3 optically couples the beam to the BOMEM MCT detector. The entire optical path is evacuated thus reducing the effects of absorption by atmospheric H2O and CO2.
Sample Cell Design
The sampling cell used for acquiring low vapor pressure reference spectra was designed with the following requirements. The cell had to be equipped with windows transmissive in the 500- 4,000 cm-1 spectral range. This range covers the primary IR absorption features of the compounds of interest. Next, the cell had to function over a temperature range of 70- 500 F to provide reference spectra for all required sample conditions. The cell had to be equipped with evacuated tubes extending from the sample cell windows so the IR beam could be traversed to and from the sample cell unattenuated by atmospheric gases. Heated sample inlet and mixing chamber apparatus were required for vaporization of the low vapor pressure compounds and for sample dilution to the required ppm level. The cell had to be leak free from 0.01 to 1000 torr pressure for periods up to 40 minutes to allow sufficient time for sample preparation and spectral data acquisition. Additionally, the system had to be equipped with instrumentation to accurately determine temperature and pressure in the mixing chamber and sample cell during sample preparation and spectral data acquisition. Finally, the cell had to be designed with expandability in mind. It is anticipated that a heated multi-pass absorption cell will be required for future reference spectra measurements at lower ppm levels. To obtain the required optical configuration it was essential to locate the high temperature cell and associated hardware external to the FTIR. The optical layout of the FTIR measurement system is illustrated in Fig. 1. A 4 ft. by 12 ft. stable table was used for mounting the boxes and tubing used in the optical configuration of the high temperature sample cell. The BOMEM FTIR was attached securely to the stable table to minimize optical beam deviations when the external optical boxes are evacuated.
A quartz sample cell previously used at AEDC to measure the optical properties of CO, CO2, H2O and NO at temperatures as high as 1000 K was selected. The only change was the use of zinc selenide windows to obtain the required optical transmission in the 500- 4,000 cm-1 range. The quartz cell was chosen because of its inertness at the elevated temperature. Dimensions of the sample cell were limited by the size of the furnace used to heat the sample cell. A Mellen (Model No. 2-301-2) three zone clam shell furnace was selected for this purpose. The dimensions of the furnace are 3 in. in diameter by 13 in. in length. To insure uniform heating, the sample cell was limited to 5 in. in length. Additionally, three zones of temperature control were used to regulate temperature of the clam-shell heater. Kalrez was selected for the O-ring seal material because of its high temperature softening point and inertness.
The sample cell design is shown in Fig. 2. Cell windows were secured by pressure applied to three spring-loaded rods shown in the figure. The rods and flange assemble were constructed of 316 stainless steel. Kalrez o-rings were also located on the window face opposite the cell. These rings sealed the evacuated tubes extending from each end of the cell. Belleville spring washers, located outside the furnace were used to maintain constant pressure on the cell window seals. With the springs outside the furnace an initial tightening of the nuts of 80 in. lb. was sufficient to maintain a tight seal while the temperature was increased. The sample cell is shown mounted in the clam shell heater in Fig. 3.
The sample cell was connected to the gas handling system via the quartz fill tube illustrated in Fig. 2. All components of the gas handling system including the sample inlet, mixing chamber, valves, transfer lines and pressure transducers were made of 316 stainless steel. Where possible, stainless steel surfaces in contact with the sample were gold plated to assure inertness. Fig. 4 depicts the components of the gas handling system. This system was wrapped with heat tape in four separate zones for temperature control. Three additional type K thermocouples were installed to monitor system temperature. An OMEGA model CN3390 multiloop temperature controller was used to regulate temperature of the individual zones of both the quartz sample cell and sample handling system. Provisions were made for in-place calibration of the heated pressure transducers using the reference pressure transducers shown in Fig. 4.
The integrated FTIR, sample cell and gas handling system are illustrated in Fig. 5. Three individual pumping systems, each equipped with separate vacuum pumps were included. The sample cell and gas handling system also contain a turbomolecular pump to provide the vacuum required for decontamination. One pumping system was dedicated to the BOMEM FTIR instrument. The second pumping system was used to evacuate the optical boxes and tubing through which the beam traverses to reach the sample cell and detector. This pump-down capability provides an absorption free path through this section of the optical system, thus providing reference spectra with a background spectrum (Io) unattenuated by atmospheric gases. A third vacuum system used to pump down the sample cell and gas handling system provides the high vacuum required for system decontamination and sample preparation at the required species concentration.
Sample Preparation and FTIR AnalysisOur plan for obtaining the reference spectra calls for acquisition of duplicate spectra for each compound at three path integrated concentrations in the range of 20- 500 ppm-meters. Using the gas mixing apparatus described above, only a few grams of each low vapor pressure compound were needed for this purpose. To prepare the calibration gas standard a small quantity of the compound was placed in the sample holder shown in Fig. 4. The sample holder was immersed in liquid nitrogen (LN2) and the gas handling system and quartz cell were pumped down to a vacuum. This process was performed to remove impurities trapped in the gas handling system and the sample cell. After a thorough pump- down, the sample was vaporized by increasing the temperature of the sample holder. During this time the quartz sample cell was adjusted to the required temperature for acquisition of the high temperature spectra. Next, the vaporized sample was introduced into the mixing chamber and sufficient time was allowed for stabilization before recording temperature and pressure. The sample contained in the mixing chamber was diluted with preheated ultra-pure nitrogen to approximately 900 torr pressure. This pressure was sufficient to fill the sample cell to the required 760 torr pressure for acquisition of the reference spectra. Typically, 0.2- 5 torr sample pressure in the mixing chamber was sufficient to provide the required absorption level at a total pressure of 760 torr in the quartz sample cell.
RESULTSApproximately thirty low vapor pressure compounds requiring high temperature reference spectra were initially identified by the EPA. Half of these compounds have been procured and work has begun on acquiring their spectra. Three mercury compounds (Dimethylmercury, Methylmercuric Chloride and Mercury(II) Chloride) were selected for initial reference spectra measurements. Preliminary absorption spectra of two compounds (dimethylmercury and methylmercuric chloride) have been obtained at this time.
Absorption spectra for these organomercury compounds were acquired at 0.125 cm-1 resolution and processed for the reference spectra database at 1.0 cm-1 resolution. Where possible, duplicate spectra were obtained at three or more concentrations and three temperatures. Typical results for dimethylmercury and methylmercuric chloride spectra are presented in Figures 6 through 9. In Fig. 6 the ambient temperature absorption spectrum of dimethylmercury at 404 ppm-meter concentration is presented. In this broadband spectrum one can see several absorption bands with unique line structure for identification of the compound. In Fig. 7 the absorption spectra of dimethylmercury at 270 ppm-meter concentration is shown for three temperatures (80, 150 and 250 F). This figure shows the CH vibration band of the compound at 3000 cm-1. In the figure one can see some differences in the three spectra. Of particular interest are several additional lines that show up in the higher frequency region of the 250 F spectrum. Many of the lines observed in this bands coincide with absorption lines of methane, suggesting that a methyl group may have been stripped off the compound in the process of preparing the high temperature gas sample. Evidence of this is presented in Fig. 8 by comparing the absorption spectrum of dimethylmercury with that of methane. The methane lines were uncovered during a spectral search of the dimethylmercury spectrum using the QASpec Spectra Library published by Infrared Analysis, Inc. of Anaheim, CA.
Due to the low vapor pressure of methylmercuric chloride we were unable to obtain a reliable ambient temperature spectrum of the compound. However, duplicate spectra were obtained at 150 and 250 F. The 250 F spectrum at 149 ppm-meter concentration is presented in Fig. 9. Similarly, in this spectra there are several unique features present that look promising for identification of the compound. It is also believed that this compound decomposed partially at the elevated temperature. Evidence of this is present at the 1,300 and 3,000 cm-1 regions of the spectrum. Many of the absorption lines in these bands also coincide with absorption lines of methane (CH4).
With little information available on the stability of mercury compounds at elevated temperatures and no supporting information on purity of the vaporized sample, the results presented here are considered preliminary. Future plans call for the use of a gas chromatograph to measure purity of the vaporized sample at the time of sample preparation and analysis.
SUMMARYFourier Transform Infrared (FTIR) spectroscopy is a technique with the potential to detect and quantify many of the hazardous air pollutants identified in Title III of the Clean Air Act of 1990. We have reported on the status of an experimental program sponsored by the EPA's Office of Air Quality Planning and Standards to establish reference spectra for many of the low vapor pressure hazardous air pollutants. This information will be used by the EPA in their efforts to establish maximum achievable control technology standards for stationary sources. Fabrication of the apparatus required for preparation of high temperature calibration gas standards is complete and the equipment has been incorporated into a high resolution FTIR Spectrometer system. This system provides quantitative calibration gases in the temperature range of 70- 500 F and spectral coverage from 500- 4,000 cm-1. An experimental program to prepare reference spectra for thirty or more of the low vapor pressure compounds is underway. The absorption spectra of two organomercury compounds (dimethylmercury and methylmercuric chloride) have been obtained and are reported here. Since evidence suggests that the compounds may have partially decomposed at the higher temperature spectral results obtained thus far are considered preliminary.
ACKNOWLEDGEMENTSThe authors would like to thank Drs. Grant M. Plummer, Thomas M. Geyer and Thomas A. Dunder of Entropy Environmentalists, Inc. for sharing their experiences obtained while measuring the FTIR reference spectra of over 100 other HAP's. We would also like to thank all the craftsmen and technicians who provided assistance in fabrication and assembly of the experimental hardware used in this study.
2. G.M. Plummer and L.T. Lay. "Technical Aspects of EPA's FTIR Development Projects," presented at 85nd Annual Meeting of A&WMA, Kansas City, Missouri, June 1992.
3. W.J. Phillips, J.H. Welch and B.J. Brashear. "A High Temperature Infrared Absorption Gas Cell," Review of Scientific Instrumentation, Vol. 63, No. 4, April 1992
2. S.S. Penner, Quantitative Molecular Spectroscopy and Gas Emissivities; Addison-Wesley: Reading, MA, 1959