THE UNIVERSITY OF MICHIGAN COLLEGE OF ENGINEERING High Altitude Engineering Laboratory Departments of Aerospace Engineering Meteorology and Oceanography Quarterly Report HIGH ALTITUDE RADIATION MEASUREMENTS 1 January 1970-31 March 1970 Fred L. Bartman ORA Project 03635 under contract with: NATIONAL AERONAUTICS AND SPACE ADMINISTRATION CONTRACT No..NSR 23-005-376 WASHINGTON, D. C. administered through OFFICE OF RESEARCH ADMINISTRATION ANN ARBOR April 1970

_e_ a 4c&-, '!^ &crvy s\jvbYoE3~

Table of Contents Page List of Illustrations Abstract I. Introduction II. High Resolution Measurements cfthe 15gm Absorption Band of CO2 (L. W. Chaney). A. Status of the Jarrell Ash Spectron e Beginning of the Reporting Period. B. Tasks Completed During the Reporting Period C. Detailed Discussion of Work Carried Out During the Reporting Period. III. Theoretical Analysis of High Resolution 15tm. C02 Measurements (S. R. Drayson). IV. Development of Ozone Generating, Handling and Measuring System (L. T. Loh and P. A. Titus). A. Ozone Generating and Handling System B. Quantitative Mleasurements of Ozone V. Theoretical Analysis of 9, m Ozone Band (W. R. Kuhn). VI. Literature Survey on the Distribution of CO2 in the Earth's Atmosphere. VII. Plans for Continued Work. VIII. References v vil 1 1 1 2 2 5 5 5 6 7 8 9 10 iii

List of Illustrations Figure Page 1. Jarrell Ash 1. 8 meter spectrometer and associated experimental apparatus. 11 2. Experimental spectrum of C02 absorption showing the Q-branch of the (010:1-100:0) band (band 3). Also visible are the low J-number lines of the P and R branches of the same band and high J-number lines of the fundamental (band 1). Pure CO2 in cell, length=8. 74 cm, pressure = 100 mm, temperature =298~K. 12 3. Theoretically calculated spectrum for the same conditions as the spectrum in Figure 1. Resolution of the theoretical spectrum is 0. 06 cm-1. 13 4. Schematic diagram of ozone generating and handling system. 14 v

Abstract This report summarizes project activity during the period 1 January 1970 to 31 March 1970. Experimental work discussed includes the high resolution measurements of the 15 m band of CO2 and the development of an ozone generating handling and measuring system. Theoretical analysis of the high resolution 15/,m CO2 data and of the 9. 0 03 band are in progress. A literature survey on the distribution of C02 in the earth's atmosphere is being made.

I. Introduction This is the 2nd Quarterly Progress Report on Contract No. NSR 23-005-376, covering the period 1 January 1970 to 31 March 1970. Since the administrative decision to change this contract came after the Quarterly Progress Report for the period 1 October 1969 to 31 December 1969 had been distributed, that progress report has the number 05863-28-P instead of 03635-1-P. The project effort during this time was divided among the following tasks. 1. High resolution measurements of the 15pm absorption band of CO2 (L. W. Chaney). 2. Theoretical analysis of high resolution 15 r m CO2 measurements (S. R. Drayson). 3. Development of ozone generating, handling and measuring system (L.T. Loh and P. A. Titus). 4. Theoretical analysis of 9. 6j m ozone band (W. R. Kuhn). 5. Literature survey on the distribution of CO2 in the atmosphere. I. High Resolution Measurements of the 1512 m Absorption Band of CO9 (L. W. Chaney). A. Status of the Jarrell Ash Spectrometer at the Beginning of the Reporting Period 1. Test spectra using the 8. 74 cm cell and the gas handling system had been made. The spectra indicated that a resolution better -1 than. 08 cm could be achieved. Also, the signal to noise ratio was satisfactory (see spectra in previous report). 1

2. The components required to interface the spectrometer and the IBM card punch were designed and complete, but not assembled or checked. B. Tasks Completed During the Report Period 1. The spectrometer and the IBM card punch were connected. 2. A new interference filter was installed on the dewar helium station. 3. An investigation was made of the instrument resolution. 4. A wavelength calibration of the instrument was made. 5. Thermistors were installed to monitor the instrument temperature. 6. A nitrogen purge system was provided for the source area to reduce CO2 background. C. Detailed Discussion of Work Carried Out During the Reporting Period. Originally it was planned to use a G. P. 2402 A integrating voltmeter (61 second integration time) and a H. P. 2547 A coupler to interface the spectrometer with the IBM summary punch. Since both items were available at the Willow Run Laboratory this was a very economical plan. However, further analysis showed that a better signal to noise ratio could be obtained with a system using the High Altitude Engineering Laboratory H. P 2410 C integrating voltmeter and an external oscillator to increase its integrating time from 1 second to 2. 2 seconds. Since the output logic of the two voltmeters was different, two new cards were required for the coupler. They were purchased. A new interference filter, used to isolate the first order of the grating was purchased from Optical Coating Laboratories Inc. The only available filter in stock was 1, 5 inches in diameter. It was cut into two pieces. One piece was 0. 75 inches diameter and the other 0. 562 inches in diameter. The larger 2

piece was mounted in a window holder with a KBR window. The purpose of placing the filter in the window was to reduce the thermal strain due to temperature cycling the detector. However, as a result of the increased background photons on the detector the sensitivity was reduced by 50%. The smaller piece was mounted on the helium station and the original sensitivity was restored. Thus far, there has been no discernible deterioration similar to that noted for the original filter. The fact that the new filter is more than twice as thick as the old one (0. 018-.040 inches) maybe significant. Also, less epoxy was used in cementing the filter to the helium station. After completion of the card punch system and installation of the new filter, test spectra were run to check the entire system. The measured spectra were compared with calculated spectra in the region of 720 cm (see figures 1 and 2) and near 660 cm The following conclusions were made as a result of these measurements. 1. The wavelength calibration was not linear. 2. The resolution varied between.055 and. 07 cm 3. The wavelength markers shifted with room temperature. 4. A method was needed to synchronize the initial card punch with a given wavelength marker. 5. Small amounts of CO2 appeared in the background scans. A very careful wavelength calibration of the anticipated spectral range was carried out. Several attempts were made to correct the non-linearity of the wavelength drive by adjusting the optical flat used in the sine bar drive. After several attempts to eliminate the non-linearity it became apparent that the maximum possible correction was inadequate. Hence, the sine bar was returned to the nominal position and a wavelength correction curve was generated. 3

The daily shift in the correction curve seems to be linear and dependent on room temperature. A single point will be used to re-check the calibration prior to recording a data set. The individual lines in the Q-branch at 720 cm vary in spacing from -'1 0. 15 to 0. 01 cm. Hence, a very small change in resolution can be detected in this region of the spectrum. A comparison between the measured spectrum and the theoretical spectrum reveals very clearly the instrument resolution. Many spectra were taken of the same spectral region and it was determined that the change in resolution was related to the room temperature changes. A resolution less than 0. 06 cm can only be obtained when the variation in instrument temperature is less than 0. 25 C. The normal ex-1 pected resolution will be between 6. 06 and 0. 07 cm. The change in resolution with variation in temperature across the instrument is due to distortion of the optical mounts in such a way that the image of the slit rotates slightly. Three thermistors were placed on the instrument to monitor temperature. The thermistor readings will be a part of the housekeeping data associated with each scan. Synchronization of the card punch with wavelength markers is accomplished by aligning a pointer on the marker disk with a corresponding pointer on the photo cell housing using a large mirror which is placed behind the wavelength drive. The punch is started when the pointers are alined. The entire area between the source and the input optics has been sealed with masking tape and purged with nitrogen gas. There is now no indication of CO2 in the background spectrum. The spectrometer is now completely ready for data recording. Instrument problems which have been postponed to a later date are: the stray light, fore pump vibration, and sealing of the main chamber. The instrument and associated apparatus are shown in figure 1. 4

III Theoretical Analysis of High Resolution 15p m, CO2 Measurements e. R. Drayson) The analysis of the data is still in a preliminary stage, but is expected to proceed more rapidly in the near future since most of the preliminary instrument work is now completed. A computer program was written to convert the spectrometer card output to absolute transmittances including computer produced plots of the spectra. At the same time existing programs were modified to calculate theoretically the absorption under the same conditions as the experimental spectra. Typical results are shown in figures 2 and 3. These comparisons are useful in estimating the resolution of the experimental data, as well as checking on frequency calibrations and making rough estimates of the accuracy of the input parameters used in the theoretically generated spectra. IV Development of Ozone Generating, Handling and Measuring System (L. T. Loh and P. A. Titus) A. Ozone Generating and Handling System. Figure 4 is the schematic diagram of the ozone generating and handling system. The pipes which connect the oxygen and nitrogen tanks to the flowmeter, ozonizer and 40 meter cell "S" have been installed. The connection of the metal tubing to the glass tubulations of the ozonizer were critical. The metal tubing has been anchored to the ozonizer wooden box and connections to the glass tubulations were made by Tygon tubing with Teflon seals. Pump out of the used sample has been arranged to be made directly into the pump or through a molecular sieve for destruction of 03. Two pumps have been installed in parallel. The Welch pump is fast, but its oil is incompatible with 02 or 03. The small "Will" pump has a case of stainless 2 o3 steel and working mechanism of neoprene. It should be quite satisfactory for low concentrations of 03 in 02. 5

Pressure gauges suitable for use with 03 have been ordered. A Datametrics Barocel Electric Manometer, Type 1023, with a 2000 mm Hg full scale maximum range and a 2 mm. Hg fullscale minimum range has been procured. A dial type pressure gauge made of 316 stainless steel will also be ordered. A neon sign transformer which was on hand for use with the ozonizer was found to be defective. A new one was ordered, it is a special design with one side of the secondary winding grounded. A pressure filter containing Linde 13-X molecular sieve material will be used to destroy ozone. Leaks in the O ring seals of the pressure must be eliminated before this device can be used. The 0 rings on the 40 meter gas cells have been coated with Ascolube (Fluorositirone) grease so that they can be used with the 03 mixture at low pressures. Eventually they will have to be replaced. A study of O ring materials and design has been made. Teflon and viton are materials best suited for use with 03, however teflon has low resilience and viton is low in tear resistance. Both materials have a tendency to cold set. Teflon coated rubber O rings are also available. This combination combines the resilience of rubber with the excellent chemical resistance of teflon. This last type of O ring has been ordered for use with our system. B. Quantitative Measurements of Ozone A check of commercial instruments which can be used for quantitative measurements of ozone was made. The search indicated two instruments which would be quite suitable for this purpose, a Jarrell Ash 1/4 meter Ebert, model #82410 and a Heath Czerny Turner model EU-700. Although the Heath instrument had many intersting up to date design features deemed to be very desireable, delivery was uncertain. The Jarrell Ash Instrument has been ordered. 6

V. Theoretical Analysis of 9 H m Ozone Band (W. R. Kuhn). For remote sensing purposes, it is necessary to know the spectroscopic band parameters (line strengths and positions) quite accurately. Recent work being done at the University of Michigan utilizing the9ma spectral region of ozone indicates that there is need for more accurate parameters 1 for this band. Russell (1970) has found that uncertainties in these parameters in the band wings can lead to appreciable errors in ozone profiles and amounts. Also spectroscopic observations reveal discrepancies between transmission functions and theoretically calculated profiles (Drayson and 2 Young, 1967). While our present knowledge of the m region may be adequate for flux divergence calculations for climatological purposes, more accurate calculations are necessary for remote sensing investigations. Although band parameters for the!Qmregion have been determined by 3 Clough and Kneizys (1965), there are discrepancies between transmission profiles calculated using their data and experimental measurements; there is thus a need to repeat these calculations. To gain familiarity with the types of calculations involved, line positions for the! (1103 cm ) and 3 (1042 cm ) transitions are being calculated for a simple asymmetric rotor. The rotational constants used are from Clough and Kneizys (1965). The energy matrix is factored by the Wang transformation into the four submatrices and the roots (energy levels) of the characteristic equations of r these submatrices and then evaluated. The type I representation was chosen since for nearly prolate asymmetric rotors (K -0. 968 for ozone), the energy matrix is nearly diagonal and leads to the mrost rapid solution. The roots are approximated by the continued fraction form of the submatrices (Allen 4 and Cross, 1963). The importance of centrifugal distortion is demonstrated if one compares the rigid rotor model to the quantum mechanical results of 7

Clough and Kneizys (1965). For the rotational level 15 11 in the /3 -1 vibrational state, we find an energy of 1193. 36 cm while the more precise treatment of Clough and Kneizys gives 1190. 96 cm. Differences in ground state energies for the two methods are much less, e. g. the rigid rotor energy for the 154 12 rotational level differs by only 0. 28 cm from the quantum mechanical result. The inclusion of centrifugal distortion and coriolis effect will be individually checked and the influence of each of these on the transmission functions will be systematically decuced. These calculations will make use of and be compared with ozone measurements which will soon be made at the High Altitude Engineering Laboratory. Improved band models will also be developed for use in atmospheric flux divergence calculations. VI Literature Survey on the Distribution of CO2 in the Earth's Atmosphere The literature survey of CO2 in the atmosphere has been continued. A large portion of the recent scientific activity is concerned with the seasonal variation of CO near the earth's surface. The recent results on CO2 in the upper atmosphere have been outlined in the last 2quarterly reports, 05863-27-P and 05863-28-P. Only one new brief note on CO2 in the atmosphere away from the 5 earths surface is that by Cadle and Allen5 'Present estimates of CO2 concentrations near ground level average about 330 ppm. Although there is some evidence that it decreases by about 5 ppm above the tropopause, recent measurements made by Scholz, et al (33) in the vicinity of the stratospause indicate a CO2 concentration of about 320 ppm. A bibliography of references on CO2 in the earths atmosphere is being prepared. 8

VII Plans for Continued Work Work during the period 1 April - 30 June, 1970 will include the following. A. High Resolution CO2 Measurements: Spectra of the R branch of the fundamental (R6-R10, R36-R40,R54-R60) will be recorded using an 8. 74 cm. cell and with pure CO2 at eight pressures in the range of 5 mm. Hg. to normal atmospheric pressure. An investigation will also be made of the Q branch corresponding to the isotope C12016018 near 662 cm. After the above spectra have been studied and analyzed, new regions yet to be selected will be studied. Other cells 10. 5 cm, 2. 5 cm and 5 cm) will be used in addition to the 8. 74 cm cell. If it should be decided that stray light in the instrument is intolerable, masking modifications will be made. B. Ozone Generating, Handling and Measuring System The work to be carried out on the construction of this system is listed in detail below. Estimated completion dates for each phase of the work, are given. 1. Checkout Jarrell Ash Monochrometer mercury lamp and photomultiplier May 1 2. Complete ozonizer exhaust system May 1 3. Complete Barocel piping May 8 4. Complete molecular sieve piping May 8 5. Complete breadboard of U, V. transfer optics May 8 6. Leak check ozonizer system May 15 7. Test run Perkin-Elmer 221 Spectrophotometer and oxonizer May 31 8. Design adapter for transfer optics, fabricate adapter and install adapter June 1 9. Complete stirring device for sample cell June 1 10. Test run UV 03 analyzer June 16 11. Take 03 spectra June 30 9

C. Theoretical Analysis The analysis of the C02 R branch spectra will be carried out. Studies of the band parameters of 9,m 03 will be continued. D. Distribution of CO2 in the Earth's Atmosphere A report, containing a bibliography and summary comments on CO2 in the earth's atmosphere, will be written. VIII References 1. Russell, J. M., "Measurement of O0 from Satellite Infrared Observations in the 96,um Band"1, PhD Dissertation to be published by summer 1970. 2. Private communication 3. Clough, S. A. and Kneizy's, F. Y. "Ozone absorption in the 9 Micron Region, " Report, AFCRL 65-862, November 1965. 4. Allen, H. C., and Cross, P. C., "Molecular Vib-Roters, John Wiley and Sons, 1963, 323 pages. 5. R. D. Cable and Allen, E. R., "Atmospheric Chemistry," Science vol. 167, no. 3916, 16 Jan. 1970, pp 243-249. 10

BARATRON. PRESSURE GAGE TEXAS INSTRUMENT Cu:Ge DETECTOR I JARRELLASH 1.8 METER SPECTROMETER ANALOG RECORDER F.~P — DIGITAL OUTPUT PUNCH CARDS Figure 1. Jarrell Ash 1. 8 meter spectrometer and associated experimental apparatus

f-r,4 Figure 4. Schematic diagram of ozone generating and handling system.

UNIVERSITY OF MICHIGAN 3 90II11111111111115 02227 11 601111111llll 3 9015 02227 1160