ENGINEERING RESEARCH INSTITUTE UNIVERSITY OF MICHIGAN ANN ARBOR Technical Note COMBUSTION STUDIES WITH A 12000-CURIE GOLD SOURCE Control No. AFOSR-TN-56-17 By Stuart W. Churchi'l1 Alexander Weir, Jr.; Leroy:'F. OrneIa;. " ~" Roy:. L "Gealer: R' bert E. Kelley. Martin E. Gluckstein." ~.. Project 2288 DEPARTMENT OF THE AIR FORCE, AIR RESEARCH AND DEVELOPMENT COMMAND CONTRACT AF 18(600)-1218, AF PROJECT 37507 December, 1955

00*q> 3oes

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN PREFACE This research was supported by the United States Air Force through the Air Force Office of Scientific Research, Air Research and Devetlopment Command. Beside the authors of this report, the following personnel of the University of Michigan contributed to the research work. Mr. A. H, Emmons of the School of Public Health helped plan the shielding and irradiation, assisted in monitoring the source durs ing the experimental work, and calculated the rate of irradiation of the flame and fuel-air mixture. Mr. C. C. Palmiter and Mr. W. R. Dunbar of the School -of Public Health directed the transfer of the source from the Materials Testing Reactor to the laboratory and monitored the source during the experiment, Dr. J. H. Enns of the Department of Physics helped realign the spectrograph.. Mr. T. B. Khammash of the Aircraft Propulsion Laboratory assisted in the calculation of rotational temperature. Mr. M. P. Moyle of the Aircraft Propulsion Laboratory assisted in installing the source and in analyzing the fuel-air mixture. Mr. R. E. Cullen of the Aircraft Propulsion Laboratory assisted in the installation of the source. Mr. E. T. Howard of the Aircraft Propulsion Laboratory assisted in the experimental work. The invaluable advice and assistance of the following non-university personnel are gratefully acknowledged. Personnel of the Selfridge Air Force Base transported the gold to and from the Materials Testing Reactor. Captain Leroy L. Lette, USAF, of the Office of Scientific Research and personnel of the 30th Air Division (Def) helped expedite the transportation arrangements. Dr. W. B. Lewis of the Atomic Energy Division, Phillips Petroleum Co., helped ascertain the strength of the irradiation of the gold. iii

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN ABSTRACT The effect of nuclear radiation from Au-198 and Au-199 on propaneair flames at low pressures was investigated experimentally. The source had an initial strength of 15,270 curies. Both the pre-flame mixture and the flame zone were irradiated. The number of ion pairs produced by beta radiation greatly exceeded that produced by gamma radiation. The rate of propagation of bunsen flames was increased up to 50 per cent by irradiation. The emission due to CH, C2, and OH was measured at a series of elevations through a flat flame, and CH rotational temperatures were computed. The emission at 4315 A due to CH was increased up to 33 per cent, and the emission at 5165 A due to 02 was increased up to 150 per cent with the source at a level of 1500 curies. The emission due to OH was not affected significantly. Increases in CH rotational temperature up to 100 per cent were calculated for a source level of 1500 curies. The maxima in the emission due to C2 and OH were observed to occur at an earlier stage than the maximum in emission due to CH in the irradiated flames................ iv

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN TABLE OF CONTENTS Page LIST OF FIGURES vi PART I. INTRODUCTION 1 PART II. EXPERIMENTAL EQUIPMENT 1 A. Burner Assembly 1 B. Fuel-Air Supply 4 C. Vacuum System 4 D. Optical System 4 E. Shielding 5 PART III. SOURCE 6 A. Selection 6 B. Preparation 6 C. Transfer 6 D. Activity 11 PART IV. EXPERIMENTAL PROCEDURE 14 A. Time of Experiments 14 B. Fuel-Air Ratio 14 C. Flame-Speed Experiments 14 D. Spectrographic Traverses 18 PART V. RESULTS 19 A. -Flame Speed 19 B. Spectral Intensities 22 CH at 4315 A 22 C2 at 5165 A and OH at 3063 A 23 PART VI. DISCUSSION 23 A. Flame Speeds 23 B. CH Rotational Temperatures 33 Procedure for Computing Rotational Temperature of 33 CH Molecules Computed CH Rotational Temperatures 37 C. Spectral Intensities 41 PART VII. SUMMARY AND CONCLUSIONS 44 APPENDIX 47 BIBLIOGRAPHY 53 vr..

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY. OF MICHIGAN LIST OF FIGURES Figure Page 1 Experimental equipment 2 2 Burner assembly 3 3 Disintegration schemes of Au-198 and Au-199 7 4 Aluminum cylinder for retention of gold in Materials Testing Reactor 8 5 Shipping container g 6 Lowering the shipping container into tank (rehearsal) 10 7 Installing the burner head (rehearsal) 10 8 Source strength during experimental program 13 9 Typical photograph of irradiated bunsen flame 15 10 Sample microphotometer tracing 20 11 Area measurement for flame-speed calculations 21 12 Effect of fuel-air ratio on flame speed of irradiated flames 24 13a The effect of radiation on flame speed 25 13b The effect of pressure on flame speed of irradiated flames 25 14 Emission at 4315 A within irradiated flames (F/A = 0.0802 - 0.0883) 26 15 The effect of radiation on maximum emission at 4315 A (F/A = 0.0802 - o.o883) 27 16 Emission at 4315 A within irradiated flames (pressure = 14 in. Hg) 28 17 The effect of fuel-air ratio on maximum emission at 4315 A in irradiated flames (pressure = 14 in. Hg) 29 18 Effect of radiation on maximum emission at 4315 A (pressure = 14 in. Hg) 30 19 Emission at 5165 A and 3063 A within irradiated flames (F/A = 0.0802 - o.o883) 31 20 The effect of radiation on maximum emission at 5165 A and 3063 A (F/A = 0.0802 - o.o883) 32 21 P, Q, and R branches of a microphotometer tracing 34 22 Microphotometer calibration 35 23 Spectral intensity as a function of quantum number 38 24 CH rotational temperature within an irradiated flame 39 25 Effect of radiation on CH rotational temperature 40 26 Comparison of emission at 4315 A, 5165 A, and 3063 A within irradiated flames (pressure = 14 in. Hg, F/A = 0.0802 - 0.0883) 42 27 Effect of radiation on location of maximum emission at 4315 A, 5165 A, and 3063 A 43 vi

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN BIBLIOGRAPHICAL CONTROL SHEET 1. Originating agency and monitoring agency: O.A.: University of Michigan, Ann Arbor, Michigan MA.: Office for Advanced Studies, Air Force Office of Scientific Research 2. Originating agency and monitoring agency report numberO.A.: ERI Technical Note 2288-6-T M.A.: AFOSR Technical Note 56-17 5. Title and classification of title: COMBUSTION STUDIES WITH A 12000-CURIE GOLD SOURCE (UNCLASSIFIED) 4. Personal authors: Stuart W. Churchill, Alexander Weir, Jr., Leroy F. Ornella, Roy L. Gealer, Robert E. Kelley, Martin E. Gluckstein 5. Date of report: December, 1955 6. Pages: 60 7. Illustrative material: 27 figures 8. Prepared for Contract No.: AF 18(600)-1218 9. Prepared for Project No.: TASK 37507 10. Security classification: UNCLASSIFIED 11. Distribution limitations: None 12. Abstract- The effect of nuclear radiation from Au-198 and Au-199 on propane-air flames at low pressures was investigated experimentally. The source had an initial strength of 15,270 curies. Both the pre-flame mixture and the flame zone were irradiated. The number of ion pairs produced by beta radiation greatly exceeded that produced by gamma radiation. The rate of propagation of bunsen flames was increased up to 50 per cent by irradiation. The emission due to CH, C2, and OH was measured at a series of elevations through a flat flame, qnd CH rotational temperatures were computed. The emission at 4315 A due to CH was increased up to 33 per cent, and the emission at 5165 A due to C2 was increased up to 150 per cent with the source at a level of 1500 curies. The emission due to OH was not affected significantly. Increases in CR rotational temperature up to 100 per cent were calculated for a source level of 1500 curies. The maxima in the emission due to C2 and OH were observed to occur at an earlier stage than the maximum in emission due to CH in the irradiated flames. vii

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY..OF MICHIGAN PART I. INTRODUCTION In the course of investigations of the utilization of gross fission products at the University of Michigan, flames were irradiated with gold and palladium sources. These investigations, which were carried out at low intensities, indicated some effect upon flame stability (1) and no effect upon flame speed (2, 3). The objective of this investigation was to study the effect of more intensive radiation upon flames. Previous studies on ramjet burners and..low-pressure flames indicated that the emission spectra of the flame provide more information on the mechanism of the combustion process than flame speed and stability measurements (4, 5). Accordingly, equipment was designed to obtain a spectral traverse of a low-pressure flat flame subjected to nuclear radiation. The intensity of the C2, CH, and OH bands was measured, and CH-rotational temperatures were computed as a function of distance through the flame at several levels of radiation. During the course of the experiment itself, it was found possible to obtain flame-speed measurements as well as the spectral data. The source of radiation was approximately 30 grams of gold wire with an intensity of about 15,000 curies at the beginning of the experimental work. A series of measurements was made as the source decayed to 0.oo0008 of its initial intensity. PART II. EXPERIMENTAL EQUIPMENT The equipment consisted of a premixed propane-air supply and metering system, a burner designed to produce a flat flame, a vacuum tank and system, radiation shielding, an optical traversing system, and the source of radiation. A diagram of the system is shown in Figure,1. Ao BURNER ASSEMBLY The burner and source assembly are shown in Figure 2. The burner head was a converging nozzle with a 100-mesh copper screen. stretched across the B-inch I.D. opening. With such a screen, a flat flame can be obtained -1

LEAD SHIELD- 7 m MIRROR 8 MOUNT: C DIL INDICATOR', ~SLIT iGNiTOR- j ~ 3 PLEXI S WINDOW m tTO VACUUM \I O PRESSURE GAGE Z RADIOACTIVE GOLD -4 BURNER HEAD UNISTRUT SPECTROGRAPH ro MOUNT r TO PULLEY m -SPECTROGRAPH FLAMESPEED OLDE SPECTROGRAPH DRAIN OUT FUEL IN FILMHOLDER BY ---- ---:Fig. 1. Experimental equipment z ADz..LLow.LUCK Fig....Experimental equipment L-~~~~~~~~~~~~~~~~~~~~~~~~~Z

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN UPPER BASKET SCREEN BASKET SCREEN BURNER HSCREEN CLAMP RING IN THIS CLAMP STUD/ V~a~C5~/ RGION!~/~\~,~kj~;COOLING WATER GROOVE — SCREENED SNAP-RING CHANNEL TO GROOVEBURNER TUBE COOLING WATER OUT -COOLIG WATER IN TO PACKING GLAND Fig. 2. Burner assembly

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY;OF MICH'IGAN at pressures up to one-half an atmosphere as compared to about one-tenth of an atmosphere without a screen. At higher pressures, the familiar bunsen cone is obtained. As indicated in Figure 2, the source was contained in a tapered hollow cylinder covered at the top and bottom with 100-mesh copper screen. The screen at the bottom was mounted in a ring which snapped into a groove inside of the cylinder or basket. The source was prepared from 0.005-inch diameter 99.95 per cent pure gold wire. Before irradiation, the wire was wound into a number of 1/16-inch-diameter helices. After irradiation, these helices were inserted in the basket and compressed by installation of the bottom screen. Cooling water was circulated through an annular space in the top of the 8-inch-long burner tube. The basket was supported by the tapered inner wall of the tube, and the burner head was threaded onto the outside of the tube. B. FUEL-AIR SUPPLY Five mixtures of air and propane were prepared and stored in cylinders. The cylinders were buried at some distance from the test stand for reasons of safety. The gas from the cylinders was passed through a reducing valve and a rotameter to measure the rate of flow. The rate was controlled by a hand-operated needle valve. The fuel-air mixture then passed through a flame trap and into the burner tube. C. VACUUM SYSTEM The burner assembly was located inside a tank 2 1/2 feet in diameter and 9 feet high. The removable tank cover was fastened with swing bolts and sealed with a lead core o-ring. The exhaust gas from the flame passed through a 2-inch flexible metal hose leading from the cover of the tank to a filter box. The filter box contained a Cambridge absolute filter to remove entrained particles which might be radioactive. The exhaust gas then passed through a Nash Hytor vacuum pump (using water as a working fluid) to the atmosphere. The pressure in the tank was controlled by bleeding air into pump suction. The vacuum tank also contained a pressure tap, a lamp for illumination, and a spark plug for ignition, mounted on a swinging arm and controllable from outside the tank. D. OPTICAL SYSTEM As indicated in Figure 1, a front surface aluminum-coated convex mirror was mounted in 12-inch pipe welded on a side of the tank. The mirror 4

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN reflected light emitted by the flame out through a plastic window in the tank wall and focused the image of the flame on the slit of the spectrograph. Dry air was blown on the mirror and window to prevent fogging. A slide valve was installed in the 12-inch pipe to keep the mirror dry when the tank was filled with water. The spectrograph was a Cenco student model'with a plastic replica grating. It was mounted directly on the tank ^to prevent relative motion between the spectrograph and flame image. The construction Of'the mount is shown in Figure 1.' The back of the spectrograph was pivoted on an axle set in a pair, of pillow blocks fastened to a bed plate.' The front of the spectrograph rested on an- adjusting- screw fastened to the same bed plate. Thus, the position of the horizontal slit could be adjusted up and down to traverse the flame image vertically. A dial indicator gage, reading in thousandths of an inch; was mounted in such a way that the elevation could be recorded. Thus light emitted by the flame passed from the flame to the mirror, through the window, through the spectrograph slit, to the refle-cting spectrograph grating, and then to the photographic film where the intensities of the various wavelengths were recorded. A frame made to hold a 4 x 6-inch film holder was mounted immediately in front of the spectrograph slit in such, a position that:'the- image from the mirror fo-used on the film. Photograph's obtained "in this manner were used for flame-speed measurements. E. SHIELDING Considerable shielding was necessary to protect personnel from gamma radiation during- the installation -of the gold source and during the experimental work. It was also necessary to shield photographic film during a four-hour period in the spectrograph. A hemi-cylindrical ring of lead 7 inches thick, weighing 2500 pounds, shielded one side of the burner, as shown in Figure 1. Four hundred and. fifty 100-pound bags of Portland. cement -were stacked around the perimeter of the tank.'Portland. cement was found to be'the cheapest bagged' material on a mass basis available. About 500 pound's of lead sheet were installed at strategic positions, particularly around the spectrograph. During'the installation of the source and burner head, the vacuum tank was filled with water. The water provided trnasparent shielding and permitted direct observation and manipulations from the top of the tank. 5

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN PART III. SOURCE A. SELECTION Gold was selected as a source material because the half-life obtained upon irradiation was more convenient for the experimental work than the half-life of other available sources of beta radiation. As indicated in the disintegration scheme in Figure 3, the half-life of Au-198 is 2.69 days while that of Au-199 is 3.15 days (6). This average half-life of about 3 days allowed a 24-hour experimental run to be completed with only a 20 per cent decrease in activity. The gamma radiation which is emitted from both Au-198 and Au-199 is, of course, somewhat disadvantageous because of shielding problems and the possible effect of the gamma radiation upon combustion. B. PREPARATION The source was prepared form 99.95 per cent gold wire, 0.005 inch in diameter. The wire was wound helically on a 1/16-inch rod into coils approximately 12 inches long. The coils were then rolled into two small loose balls of 15 grams each. Each ball was then placed ini an aluminum cylinder containing a helium atmosphere. Figure 4 is a drawing of the cylinder. These aluminum cylinders were irradiated for 4 1/2 days in a high-flux section of the Materials Testing Reactor, Arco, Idaho. C. TRANSFER After irradiation, the gold wire was transferred in the Materials Test Reactor canal from the aluminum cylinders to the basket shown in Figure 2 and the loading string was attached. The basket was then placed in the shipping container shown in Figure 5. The container was flown from Idaho Falls, Idaho, to Willow Run Airport by the Air Force. The container was unloaded from the plane and after removal of the cover bolts was lowered into the tank which had previously been filled with water. Figure 6 shows this operation during a rehearsal. With the container resting on the top of the lead hemisphere, the loading string attached to the basket was strung through an eye suspended above the tank. The lid of the shipping container was then removed. The basket was next pulled out of the container with the string and, with the aid of a mirror, was lowered into the bottom of the tank. After removal of the shipping container, the basket was lifted into the burner tube with the string. The string was cut and the burner head screwed into position using extension tools. Figure 7 shows a rehearsal of this operation. The cover of the tank 6

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN Au-198, HALF LIFE = 2.69 DAYS r (0.96 MEV) y (o.41 MEV) Au-199, HALF LIFE = 3.15 DAYS X (0 j2 MEV) (o(.158 mEv),3 6v y7(0.158 ),7% 7y(0.21508 MEV),367% 7 (o0.05 MEV) Fig. 3. Disintegration Schemes of Au-198 and Au-199

'rl m Z C)m SCREWDRIVER SLOT,s AD WASHER THREADED CAP'~ m, i = cs3 ma~ It lI u.~~~0 _ I~~~~~~~~~~~~~~~~~~~~~~;E CU r~Z,,-I o:)~~~~~~~~~~~~~~~~i U) C r"'.,< I 1~6 16 8 Fig. 4~. Almnmclne ~ eeto fgodi aeil etn eco Z(

HOOK RING ~ /2"0STUDS, &. EQUALLY SPACED -C) - - - - -- - - - go~~~~~~~~~~~~r C) IL 14~~~~~~~~~ 7, I, t~~~~~~~~~~~~~~~7 2 gOOK LUGS, S~~~~~~~~~~~~~~~~~~~~~~~~ECSD N18"SET TE Fig S p c a 2~~~~~~~~~ ~m 7, NOT: CNTINE MAE F CSTLEAD ENCSEDIN1/8 SHETSTEEL rS Fig. 5. Shipping container

ENGINEERING'RESEARCH INSTITUTE *' UNIVERSITY OF MICHIGAN Fig. 6 Lowering the shipping container into tank (rehearsal) Fig. 7. Installing the burner head (rehearsal) l10

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN was bolted down and the water in the tank was forced out under air pressure. The slide valve protecting the mirror was opened and air was introduced in several places for drying purposes. The entire transfer operation was carefully monitored. As a result of considerable rehearsal, all these operations were carried out with dispatch and with an exposure of less than 100 milliroentgens for any individual. After removal of the water from the tank, intensities of about 800 milliroentgens per hour were encountered in the region of the spectrograph. This high rate was due to unanticipated scatter through the optical port And from the ceiling. Original plans called for eight twenty-minute exposures on each spectrographic film and for each individual to work in this area about 10 hours per week. Since 30 milliroentgens on the film and 300 milliroentgens per week for personnel were permissible, it was apparent that considerable additional shielding was necessary in the region of the spectrograph. One hundred fifty 100-pound bags of cement and 500 pounds of lead were added to the shielding before any experimental work was begun. D. ACTIVITY As noted previously as strong a source of beta radiation as possible was considered desirable in order to bound any possible effects of beta'radiation upon the fuel-air mixture and the flame. It appeared that as much as 25,000 curies could be handled safely in the proposed experimental program. Theoretically, 25,000 curies could be obtained;ffo'oi30 grams of gold irradiated" ih a high-flwu~.regioni:of' the'Materials Testing.Reactbr The gold wire was actually irradiated for a slightly shorter period than requested. On the basis of neutron flux calculations, the activity of the gold on removal from the pile at 11 a.m. EST, July 22, 1955, was reported by W. B. Lewis (8) to be: Au-198 9,600 curies Au-199 5, 670 curies Total 15,270 curies Experimental measurements of the radiation from the gold source were made through the water in the tank during installation. These data were compared with data taken under similar conditions with a known Co-60 source. The activity of the gold computed from these measurements agreed within 20 per cent with the above values. After completion of the combustion experiments, the gold was removed from the burner and the gamma spectrum 11

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN was measured. The ratio of Au-198 to Au-199 determined from the gamma spectrum and extrapolated back to removal from the reactor was also in reasonable agreement with the above values. Accordingly, the above values were used in subsequent calculations. Based on half lives of 2.69 days for Au-198 and 3.15 days for Au-199 the curie values of the source as a function of time are represented by the equations CAul98 = 7740 e0.00537t (1) -000and 917t (2) C u 3930 e where t = time in hours measured from the hour of the first experiment at 3 a.m. EST, July 24, 1955. The periods of the various combustion experiments are noted on a plot ofthe relative intensities in Figure 8. The number of ion pairs produced in the gas phase in various regions would appear to be the most significant phenomenon resulting from the' gold source. The decay scheme in Figure 3 was used to compute the number of ion pairs per. cubic centimeter produced by the source within the cylinder containing the source and in the region of the flame itself.-.Suitable corrections were made for scattering and absorption by the gold wire, the cylinder wall, and the retaining screens. The results are summarized in Table I. TABLE I ION PAIRS PRODUCED BY GOLD SOURCE Ion pairs/(cc)(sec)(curie of Au-198 or Aul199) Beta Production Gamma Production Total Production In Flame Front From Au=198 5.38 x 109 2.07 x 09 7.45 x lO9 From Au-199 8.00 x 10 1.03 x 9 1.03 x l09 In Cylinder Containing Source From Au-198 2.42 x 1013 3.16 x 109 2.42 x 1013 From Au-199 2.68 x 1012 1l68 x 109 2.68 x 1012 12

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN FLAME SPEED RUN 1, I/Io =I FLAME SPEED RUN 2, I/Io:0.78 \FLAME SPEED RUN 3,I/Ie 0.48 0.i1~~ XSPECTRUM RUN 1, I/IO= 0.094 0. Ou I I I.. ^\ \SPECTRUM RUN 2, I/o= 0.03 2085 I ~:I- _ RN SPECTRUM RUN 3,I I/ 00043.00 0.001 SPECTRUM RUN 5, I/Io=0.0008 00001 0 00 200 300 400 500 600 700 800 TIME (hours) Fig. 8. Source strength during experimental program 135

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN PART IV. EXPERIMENTAL PROCEDURE A. TIME OF EXPERIMENTS Figure 8 shows the relative intensity of radiation during the periods in which the spectrographic traverses and the flame speed measurements were performed. As was previously stated, supplemental lead shielding was placed on the spectrograph assembly after installation of the gold. This necessitated realignment of the spectrograph, since the flame, mirror image, slit, and grating (see Figure 1) must be in the same plane. The realignment procedure involved repetitive adjustments, each followed by a thirty-minute exposure of the photographic film.at various locations within the spectrograph. This necessary but time-consuming process resulted in the time delay apparent in Figure 8 between the initial flame speed and initial spectrographic measurements. B. FUEL-AIR RATIO The fuel used in the experiments was 99.5 per cent pure propane obtained from the Mathieson Chemical Corporation. The propane was premixed with air in eight 3-cu ft cylinders buried 5 feet underground at a distance of 50 feet from the laboratory. The cylinders were evacuated, filled with propane, re-evacuated to 2 in. Hg, and finally filled with propane to a predetermined pressure. Air was then added until the total pressure was 225 psia. After a four-hour period, samples were taken for analysis. Samples were not obtained of the air-fuel mixtures used in the first two flamespeed runs and the composition was computed from the mixing procedure. Samples of subsequent mixtures were analyzed by an Orsat-explosion technique. The calculated compositions and analyses are given in Table II. C. FLAME-SPEED EXPERIMENTS The vacuum was adjusted until the desired pressure existed within the tank. The rate of flow of the fuel-air mixture was adjusted until a bunsen flame existed. A 4 x 6-inch filmholder was placed in the filmholder frame. No shutter was used. The room was darkened and the mask was removed from the filmholder and quickly replaced. With contact Process Ortho film, this procedure yielded a satisfactory image of the flame. A typical photograph is shown in Figure 9. The rate of flow was then changed and the procedure repeated. Four different rates of flow were established at each pressure. When the rate of flow was reduced, the area of the flame front decreased but the flame speed remained relatively constant. The flame speed would be independent of flow rate if a heat sink (burner head) were not lo14

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN Fig. 9. Typical photograph of irradiated bunsen flame - 15

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN Table II COMPOSITION OF PROPANE-AIR MIXTURES Pressure Propane-to-Air Ratio in. Hg lb/lb Flame-Speed Run No. 1 10 0.08*, 0.07, 0.0637*+,0.06*, 0.055* 6 0.08*, 007*, o.o637*, 0o06*, 0.055 4 0.08*, 0.07*,.0637*, 0.06*, 0.055* Flar:ee-Speed Run No. 2' 8 o.o8, 0.07*, 0.0o637, 0.0o6, 0.055 6 0.08 0.07* 0.0637, 0.06, 0.055 4 o.o8* 0.07*, 0.0637*, 0.06*, 0.055* Flame-Speed Run (Cold) 10 0.0917-00939 0 48.0748-0.0758, 710-0.0777 o.o067-0.067, 0.0603-0.0642 8 0.0917-099, 00748-0758, 0.0939 748- 8, 710-0.0777 067-0.067, o0,603-0.0642 6 0.0917-0.0939, 0.0748-0.0758, 0.0710-0.0777 0.067-0.067, 0.0603-0.0642 0o.o917-0.o0939, 0.0748-0.0758, 0.0710-0.0777 O.067-0.067, 0.0603-0.0642 Spectral Run No.- 1 14 o.o0802-0.0883, 0.0709-0.072, 0.0645-0.0682 o0.0677-0.0679 10 o o0802-o.o883 8 0.0802-0o0883 6 0.0802-0.0883 Spectral Run No. 2 14 0.0747-0.0818, 0.o613-0.0629, 0.0637*.0621-0.o641 10 0.0747-0.0818 8 0.0747-0.0818 6 0.0747-0.0818 Composition computed from mixing procedure. Fi:'or all ot'ler mixtures, the range of compositions obtained from the Orsat-explosion analyses are given. 16 -

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN TABLE II (cont.) COMPOSITION OF PROPANE-AIR MIXTURES Pressure Propane -to-Air Ratio Run in. Hg lb/lb Spectral Run No. 3 14 0.0772-0.0805, 0.0691-0.0705,.o0596-0.0629 0.0574S0.0726 10 0.077240.0805 8 0.0772-0.0805 6 0.0772-0.0805 Spectral Run No. 4. 14 0.0834-0.0852, 0.0716-0.0788, 0.0637-0.0667 0.0523-0.0640 10 0.0834-0.0852 8 o.o854-o.o852 6 o.o834-0.o852 Spectral Run No. 5 14 o.0829-0.0874, 0o0813-0.0845, 0.0704-0.0706 o.o648-o.o668 10 o.0829-0.0o874 8 o.o829-0.0874 6 0.0829-0.0874 17

ENGINEERING RESEARCH INSTITUTE ~* UNIVERSITY OFF MICHIGAN cated near the flame. The flame speeds obtained at the four rates of flow were averaged, and only the average was considered in the correlations. The pressure in the tank was changed and the entire procedure repeated at the same fuel-air ratio. Then a different fuel-air ratio was introduced and the procedure repeated again at absolute pressures of 4, 6, 8, and 10 in. Hg. On succeeding days, at lower intensities of radiation, the entire procedure was repeated. D. SPECTROGRAPHIC TRAVERSES The desired pressure was established within the tank, and the rate of flow was adjusted until a flat flame floating above the tight screen with a very thin gap between the screen and the zone of burning was obtained. The flame was essentially a disc about 1 inch in diameter and of the order of 0.05 inch thick. The exact thickness and distance above the screen varied with the pressure in the tank and the rate of flow. This thin vertical dimension of the flame was traversed by the spectrograph. The spectrograph film pack was loaded with Royal Pan film and the slit adjusted to the desired elevation. The room was darkened, the mask was removed from the film pack, and the shutter was opened for thirty minutes. The shutter was closed, and a new strip of the film was shifted into position. The slit elevation was adjusted to the elevation in the flame at which the next exposure was to be taken, and the shutter was again opened for thirty minutes. This procedure was repeated for about eight different elevations through the flame. A maximum of fourteen exposures per film was available. The pressure in the tank was then adjusted to a new value and the rate of flow adjusted so that the volumetric rate of flow at the pressure of the tank remained constant, i.e., the linear velocity of gas at the burner head was kept constant. The rate of flow for the first run was chosen so that the flame did not become a bunsen flame at the lower pressures nor burn under the screen at the higher pressures established for subsequent runs, As a further condition, the flame had to be sufficiently intense to produce a satisfactory spectrographic negative in 30 minutes. The procedure was repeated at several pressures for one fuel-air ratio. Then the tank pressure was held constant, and the procedure repeated for several fuel-air ratios. On succeeding days, at lower intensities of radiation, the entire procedure was repeated. The spectrographic films were developed under constant conditions to ensure a valid comparison of the density of exposure at the various wavelengths. They were developed in 15 gallons of DK-50 developer, diluted 1 to l-with water, at constant temperature, agitation, and developing time. The same batch of developer was used for all of the films to guard against 18

ENGINEERING RESEARCH INSTITUTE ~ UN IVERSITY OF MICHIGAN changing developer strength. The quantity of developer used was so great that it was assumed that its strength did not change appreciable with use. After developing, the films were fed through a microphotometer which scanned and recorded the density of exposure as a function of the wavelength. A sample microphotometer tracing is shown in Figure 10. PART V. RESULTS A. FLAME-SPEED EXPERIMENTS Using the photographs of the bunsen flames, the circumferential areas of the flame surfaces were computed from enlargements of the negatives, The flame- speeds were computed from the relation Vf Q/Af (3) where vf = flame speed (rate of propagation of flame normal to flame front), ft/sec Q = volumetric rate of flow of air-fuel mixture leaving the burner head, cu ft/sec Af = 2TLRnE = surface area of flame, sq ft L = length of increment along perimeter of major vertical cross section of flame R = mean radius of element of flame surface n Figure 11 is a schematic diagram illustrating the method of flame-area measurement. Equation (3) assumes that the density of the material leaving the burner head is equal to the density of the material entering the flame front, which is the usual assumption made in computing flame speeds by the Guoy method. The average flame speeds computed from the photographs at four different rates of flow are plotted versus nominal fuel-air ratio in Figure 12 for tank pressures of 4, 6, 8, and 10 in. Hg abs. As previously stated, analyses were not obtained for the fuel-air ratios used in the first two flame-speed experiments (at relative intensities of radiation of I/Io 1 and I/I = 0.78). The shift in the maxima of the curves for the second flamespeed experiment (I/Io = 0.78) can undoubtedly be attributed to an incorrect value for the computed fuel-air ratio rather than to the change in intensity of radiation. The curves for I/Io = O in Figure 12 were obtained after the source of radiation was removed from the system. The compositions of the mixtures used in all but the first two experiments were obtained by analysis. 19

0~~~~~~~~~~~~~ EG~ ~ ~ ~~~O _ Trr T.1 ITR> T T r r XTTTTTTT f r I 1 I pi I I I I is 1 -- Y H 11 111 I... 11! 1111111111l l _ r~r lilt IIIi 11IIII~l!TUlilill~l I -IFT ^ I 1 1'1 W:- T Illll~~~lllqlilt 11 11 I III 1 ttl 1 11i11II I TE~~~~~~i;1 1 1 1~ ~ ~~~~~il 11 1E, iI, lllllll~ l111 11111 1 I L1{ I I I I L I I I~~~~~~~~ II II. 1. il ll. 11 111 l Ilitl TmlI I i X 41X 01t..... 4 t:, lulllLOlMIIIII0WkIIIIII'IIIIII II vd t I o _ dFFT l l l l.. 1 l. Illllti~~~lllltllllIIM 11 I llill 11111 1 1it Ill I 0, iJIt E tE | T I I I I L-,+3+ i H i I m I ri I I I I I I I I~~~~~~~~~~~lililil T~~~ T~iiiiiUITll lll. TIIIIIIIltIl 11111111 1 1i ~- I rec |~~~~~~~~ ~~~~~~~~ I I I r I I I Il I I I 1 Tm rTrilllllllXI TI lliil11iE1 limil |I I I I I I I lt Ft.r l ||s1 I I il 1 1 I X1 |l llll l t HI l IIIIIlH il lf II L IItIIIIL L I I I I limi I CD Y k~ t~nrtntt~mtmrtm mt -I- tI I I I I ~ I 1111 IIll Il W i'1 6 m milk 1 111 o m- I~n I I r I I III I I I ll l:11 I I I 11+ 111 ILUIIIIr IIIIIIU! LI I I Oq In I I I I I I I i 1 ll l I I e~~~~~~~~~~~~~~~~~~~~~~~~~~~ I I I hlllLI0 is I'e iii iP 8I!I g g rrS H~~~~~lilil i:W~~~~~lkillL~~~~ll~~l~~lL~~lilnl11 IdTllillil[ll 1>s S L~r I T1 rl I I I I I I I I I I 1 11XIIrIiFl11 11 l i F lLLLt7LLUlLLL1[IIl I () P wrl XH W1H +lllU~~~i~l~l~trI I I I HIM llll lllI In I DIH::)I, JO.A.I. IS ISAI~n.-:IJ. LlJ. I.LSNI H:::)IlIlIlIli3mi5 INIIII IN Hill > L w 1 1 1 |n| | | | I I | I |I I I I I I -L111l11l1il,1l'pil1 a IL' 1 I ttIIII!Ie I I 11 (D [[11111111-111111111"11111~~~~~~~~~~~~~~ I il I11 Il il El H l lllllll iiii illill } 8 I!tl III Fd IWll20111011T'T4 D~~~~~~~~~~~~~~~~~~~~~~~~~~~ii H, it "Al 1 1 1 1 1 1 I rl 1 l lr~lililililil~l~rrH~llilillililililililT i I il 1il1 litlel I S1 1 1< I I I I lrIIII~rl111111lililTTUTH~li11111111111111111 I Hill I I MI1 I+ fNILIIIIIIIIIIi I l'I I I 1111i1 illinll 1ll~ lL mTTlllll u n 0 lil 1:[ | g u I 7 _ 1X111 1 1 1I lil I IIIH I! rm lill lIfG l Ti I l;i lllllumT ~I I I llln I IEWle T Ir~irIr T ii T llllllllllTlT IlbLI mLLIT LIII I (D 111N1 IlTII IIlllllEtTTlllIF~ll lll1 HillTTDIIIII TTIDT1M I o IsT I I I 11 I 111111 IILLL~l~llil~lill~iI~lI~lilI!'.lilI~li4!til~l~m~lil!I 1 11 l l (D erl11111l;IIz - I I I I~~~rSitl it~~ Iillll I 11111 j Ilillkll S 11111111 LllQ I Ct~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Cjq I L I I I I I I lllll lil It TI llH 1 I ra I I I I I I I I HI I I i 1 1 i iil 1 l lL l lnniml i lluln I m I I l i l I I~31H2 1VV 1W 1IISU~ hl~1 1 1lllll II IIHII I I I R~~rl I I rT~i I I 1 1 1 1 r~l r ll T11TI rrnTTlrr TI rnTrmlilI~ llr l~ilt I H l4 T - I II. I Iiil1llil1TI1 _I0l~]S Im I s I T 11 r~ I I rII~tl~llIT1llll~mllffllltl lll tmirni I I I Frr rT ~r r T I TI I I llTr l I lT11 rl I I1r1111m1l rnllInET III;, ~ ~~~~~~~~~~~ I I r 1r lr m In I Il~lllr nIIHII MlllIlMI!nrmrnr:!T II HHIIIIIIH i I1 i-:s~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-.'...;.: 1.1. ti NV9!-HOIW~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 11AY3II *- 1111 HilSNIIIII HM333I1331

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN.. Fig. 11. Area measurement for flame-speed calculations 21

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN Figure 8 indicates that a third flame-speed experiment was performed at a relative intensity of radiation of 0.48. The results of this experiment are not plotted in Figure 12 but were discarded because of deformation of the screen and gas leakage between screen and burner head. The deformation of the screen occurred because the system was deliberately operated (after the second flame-speed experiment) so that burning occurred on the screen. This was necessary in order to obtain a maximum emission of light during realignment of the spectrograph. A new screen was inserted before spectrographic experiments were made, and again before the flamespeed experiment at I/Io = 0. In the latter experiment, the container and gold source were placed with an identical container and an equivalent amount of copper wire. The data presented in Figure 12 indicate that an increase in intensity of radiation yields an increase in flame speed. The maxima in the flame-speed curves were plotted versus relative intensity of radiation in Figure 13a, since it was assumed that all the maxima occurred at the same:: fuel-air ratio. Figure 13b shows the effect of pressure on flame speed at three intensities of radiation. The decrease in flame speed with decreasing pressure below a certain pressure (6 in. Hg) was previously shown by Cullen (9, 10) to be due to the action of the burner head as a heat sink. B. SPECTRAL INTENSITIES 0 CH at 4315 A As indicated previously, traces of the flame were made at five different intensities of radiation, varying first pressure and then the fuelair ratio at each intensity. The strongest emission due to CH occurred at 4315 A. The relative intensity of light emitted at this wavelength is plotted versus elevation through the flat flame at various pressures in Figure 14. Separate plots are presented for relative intensities of radiation of 0.094, 0.0043, and 0.0008. The average fuel-air ratio in these experiments was 0.005. The maximum spectral intensities with respect to elevation through the flame obtained from Figure 14 are then plotted versus relative intensity of radiation for the various pressures in Figure 15. The spectral intensity at 4315 A is plotted versus elevation through the flat flame at various fuel-air ratios and intensities of radiation at a constant pressure of 14 in. Hg abs. in Figure 16. The maximum spectral intensities with respect to elevation through the flat flame obtained from Figure 16 are then plotted versus fuel-air ratio at the various intensities in Figure 17. The fuel-air ratios are plotted as a range corresponding to the results obtained from several chemical analyses of the same sample. Figure 17 is cross-plotted in Figure 18 to yield a curve of maximum spectral intensity versus relative intensity of radiation at various fuel-air ratios. 22

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MIC'HIGAN`.0 0 C2 at 5165 A and OH at 3063 A O The most intense spectral line due to C2 occurred at 5165 A, and the most intense spectral line due to OH occurred at 3063 A.. The same procedure was followed for these spectral lines as for the CH at 4315 A. However, in some cases, the intensities of C2 and OH were so low that no information could be pained from the microphotometer traces. Spectral intensities of C2 at 5165 A and OH at 3063 are plotted versus elevation at an average fuel-air ratio of 0.085 and various pressures and relative intensities of radiation in Figure 19. Maximum spectral intensities of C2 at 5165 A with respect to elevation are plotted versus relative intensity of radiation for an average fuel-air ratio of 0.085 and pressures of 14, 10 and 8 in. Hg in Figure 20. The maximum spectral intensities of OH at 3063 A are also shown for pressures of 14 and 10 in.Hg. The emission due to C2 and particularly OH was much. less than from CH. These low intensities of light, coupled with the wide variations in fuel-air ratio in the experiments at 14 in. Hg abs. prevent any meaningful presentation of the C2 and OH intensities in the form used for CH in Figure 18. PART VI. DISCUSSION A. FlAME SPEEDS The experimental data in Figure 13 clearly indicate the marked increase in flame speed caused by irradiation of the flame. The flame speed at a relative intensity of radiation of 1 corresponding to an absolute intensity of 11 670 curies is at least 50 per cent greater than the flame speed of an unirradiated flame. The negligible effect of beta radiation on flame speed reported by Morrison, Cullen, and Weir was based on experiments performed with a 6.6-curie source (2) and again with a 36-curie source (3). Figure 13 indicates that a source considerably stronger than 36 curies is required to produce a significant increase in flame speed. The increase in flame speed may be attributed to the energy impart ted to the reacting molecules by beta particles, since Table I indicates that the number of ion pairs produced by gamma radiation is negligible in comparison to those produced by beta radiation. 23

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN T=P,= IHg Hg 3.4 3.. *., I. 3.0! 2.6.a- 1 ___ /_0.78_ (n 2.2 w I /I* 0 4 1.8 1.4 1.0 PT:4"Hg PT: 6"Hg 3.4 3.0 4.6 FIg. 2 E o - rr

32 m 3.0 2.8 rn I~~~~~~~~~~~~~~~~~~~~~l~~~~~~~~l 2.6 ~ a-c AM 2.4 b IP~~~~~~~~~~~~~~.. 22 C PPg8in. 10 in.H9 U02 _ _ _ _ - CL 0~~~~~~~~~~~~~~~ f f~~~~~~~~~~~~~~= cl1.6 1.4 v w 1.2. i co~~~~~~~~~~~~~~~~~~~~~~~~~~~~~c U.4 Q22 0 QI Q02 0.3 Q.4 0.5 0.6 0.7 Q.8 0.9 1.0 0 1 2 -Z 3 P 4 5 6 7 8 9''10~-~ 3 I/ Io- RELATIVE INTENSITY OF RADIAT'ION T TANK PRESSURE - inchws Hg Fig. 13a. The effect of radiation on Fig. 13b. The effect of pressure on flame flame speed speed of irradiated flames

m Z m 1.0 Z I/Io=0.094 I/Io=0.0043 I/Io=0.0008 0.9m PT:14in. Hg F' ~- 0.8 - __ _____ O 0.7 2 0.7 R 10in.Hg -F_=1__ nH PT:14in. Hg N /~,PT=I4in.Hg Z >I O e _ _ _ __ i t I'n 0. 6 t I z PT=E8in.HgIm PTl8in.Hg I8in.Hg cr- 0.4 -- in ~~~~~~~~~~~~~~~~~~~~~~~~T6nHg Uj ~~~~~~~PT:6in.HgP~6nH rIinH (f0.3z -J > 0 =6in.Hg w =6n7g,-': o~< 0.2 0.1 0 0.02. 0.04 0.06 0.08 0.10 0.12 0 0.02 0.04 0.06 0.08 0.10 0.12 0 0.02.0.04 0.06 0.08 0.0 01 ELEVATION THROUGH FLAME- inch-(arbitrary zero) C' Fig. 14-i. Emission at 43k15 0A within irradiated. f lame s fuel/air = 0.0802 0..0883 0.2_0.4.0 0080.0 z1 ELEVAT~~~~~IOTHOGFLM-inh(abtryzr)".

ENGINEERING RESEARCH'INSTITUTE UNIVER'SITY OF;, MICHIGAN.. 1.0 I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 0.9 O ~ ~ ~ ~ ~ ~ ~ ~ ~ o~~~~~~~ o~ ~ ~ ~~~ 1l,,' 0.7, IT 14in.Hg "'o~ _____J i.0.6 Z' z wZ Iz -J 0~5T = 10in'Hg -n - - P~:Iin.Hg - - - - __i_ Iw 04 Tm6in.Hg w "0.3 X 0.2..1 0.0001 0.001 0.01 0.1 RELATIVE INTENSITY OF RADIATION Fig. 15. The effect of radiation on maximum emission at 4315 a fuel/lair = 0.0802 - 0.0885 27

1.0 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I/Io0.09 I/I=0.085 IIo=.020 I/j=0.043 IIo=000z 0.9~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ F/A=0.0802-0.0883 ~~~~ ~~ ~ F/=.77019 F I/I 0O. 0834I/o=68520 _- 0:6 AdF __ Z ~~F/A=0.0709-0.0782 t/=001300,2 0. N F/A~~~~0.0747-0.08I9 F/A~~~~~~~~0.0692-0.0805 F/AA0.00704O0O070 z F/=00-.7 /=0.0621-0.0629 F/0.059629000629 z F/A~~~~~~~~~~~~O.08I3-0.0845~~~~~~~~~~~~~A 00230060 ~~Q~4 F/AA'00677-.06-0.0678 Ffl 6 =0.05273-0.60j 1/-00648- 0066 -MFA0O7-.6. /~.67006 F/A0.0574-0.0726 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 ELEVATION THROUGH FLAME-inch- (arbitrary zero) 0~~~~~~~~~ Fig3 16. Emission at 4315 A within irradiated flames pressure = 14 in. Hg

ENGINEERING RESEtARCH INSTITUTE.: UNIVERSITY OF MICHlIGAN 0.95 0.90 0.85 0.80 l I/Io 0.094 0.75 I/Io= 0.0385 0.70 E 0 I II "' 0.65 _' Q 0.60 I/ sou 0.0205= 0.55 z 0.50 Q 0.45. 0.34.. 0 0.30...... 0.25 29

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF i MIC Ht/AN.. 1.0 - 0.9 0.8 Z 0.7 0 I i I I I I I I F~F/=0.08. 5N~~~~~~~~~~~~ 0.6 z w Iz 0.5 - F/A=0.08 - / F/A 00 707 0.6 w -/ 0.3 0.2 0.5 F/A 0.0637 X r F~~~~~~~/A=0-06 0).2 0.: C _ _ _~~ ~~ ~~ ~ ~ ~ ~~ ~ ~~ ~~ ~ ~ ~ ~ ~ ~ ~ ~~ ~~~~~~.-.. ~ _ _ -.- - - _ _ _ 0. 0001 0.0 01 0.01 0.I RELATIVE INTENSITY OF RADIATION Fig. 18. Effect of radiation on maximum emission at 4515 0A pressure = 14 in. HgT 50

0.20 OH 363A I/1.=0.094 OH- 0631~ /I.0.00v43 OH-3063A, /I.-0. 008Z 0.151.1C 0.ICt0I =1. inmHg P=1l4irg P1fltin.- =i o~P=oi'g __ P~-~PT =100in.Hg __ z 0.05~~~~~~~~~~~~~~~~l Hg.. =,4in.Ha C 0 N045 [ 2y' 2 I/I.-o.0043 C,-5165A, I/I.0.0008 >- 0.40 ""r" z,,Z -0.35c o 0.30 w 025 _> PT 14in Hg J020 ix PT=14in.Hg m or. mn ~~~~~~~~~~~~~~~~~~~T1iHg =' P.in.HgH) 0.10 TP =1in.Hg 0.05 -i. PT Hg 0 0.01 0.02 0.03 0.04 0. 05 0.0 6 0 0.0 1 0.02 0.03 0.04 0.05 0.06 0 0.01 0.02 0.03 0.04 0.05 0.06 ELEVATION THROUGH FLAME-inch-(orbitrary zero) C) Fig. 19. Emission at 5165 A and 3063 A within irradiated flames fuel/air = 0.0802 - 0.08853 Z

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN 0.08 OH-363Z P III"I 0.06. 0.04 L I........ o =o0 — T =10 Hg -' - - 0.02 1 0.5....1 z.0. I0.4:_ 0.3...... Fig. 0. The ffect of r4 of!Hn on maxim'$m emission at 5165 ii and,06j,,, ~ 0.2 Hg.,,0.1 - PT"a"Hg 0.0001 0.001 0.01 0.1 RELATIVE INTENSITY OF RADIATION Fig. 20. The effect of radiation on maximum emission at 5165 and 3063 fuel/air = 0.0802 - 0.0883 52

ENGINEERING RESEARCH INSTITUTE' UNIVERSITY OF MICHIGANB. CH ROTATIONAL TEMPERATURES The rotational temperature of the CH molecule may be computed from the spectroscopic data previously presented. The temperature is computed from the microphotometer tracings of the experimental spectrograms. A typical example of such a tracing showing the P, Q, and R branches is presented in Figure 21. The procedure for computation of the rotational temperature is outlined below. Procedure for Computing Rotational Temperature of CH Molecules The frequency of the distinct band heads were obtained from Pearse and Gaydon (11). The wave numbers, v = 1/K, of these band heads were then determined and plotted against distance on the microphotometer plot using the 4315 A line as vo. The resulting curve is shown in Figure 22. The nonlinearity of the correlation indicates that the frequency on the spectrograms is not a linear function of distance even though a grating spectrograph was used. This probably is due to the use of the plastic replica grating. However, in the region of interest for this calculation (the region surrounding the 4315 A line) the curve can be represented by v = 23, 170 - 18.9 s (4) where v = wave number s = distance from vo on microphotometer plot. Equation (4) can be combined with the expression for the wave number in terms of the rotational quantum number K, as follows: v = 23,170 - 18.9 s = 23,170 + 29K + 0.10 K2 (5) where K is the rotational quantum number and the constants 29 and 0.10 are functions of the moment of inertia of the CH molecule. The quantum number K is a positive integer and. varies By a unity from one state to the next consecutive one. K need only be determined for the first distinct peak in the R region of the bank, since the quantum number of each of the subsequent peaks is a unit higher than the preceeding peak. The distance s from v for each of these peaks was computed and is tabulated below: 33

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF. MICHIGAN,.,-, _, _. _;,. I.. z~~~~~~ _ SAMPLE MICROPHOTOMETER PLOTi _ _ 1_-USED IN CALCULATING CH _ I-~~~~~~~~~~~~~~~~~~~~~~~~~~ _-ROTATIONAL TEMPERATURE _ —-'.'9.i_.-F._. —-, 1 1'1 I _= --.... 1 —-1 —, I....:- T ~~~~~_ I::i:.B _ —'-:_: Z~~~~~~~~~~~~~~- __,.1: i. I_ I- 1 — ~....- 1'....... I~- 11 —f — — 1 F —~~~~~~~~~~~~~~-:-:. h#XX~~~~~~~~~~~~~~~~ I... I: 1 I!- n:::.HH 1IIIII 11 H H I!:. T:, t it18-m l Ut 1AN'1' 1 1,...1 0. C') z~ I —d-t:V~0qt I I I 1 F.1'1111'1 -.3 ~ ~ ~ - 3 -/z L L-,2~~~~~~~ DISTANCE (PROPORTIONAL TO WAVE NUOMUER) Fig. 21. P, Q, and_ branches of a microphotometer tracing 5::7

m m 34,000,Z 33,000 32,000, 31,000' t_ 30,000 -~29,000,o. Z 0, 28,000 -I co,J 27,000 z W m D 26,000 25,000 24,1000 ~~~~~~~~~~~~~~~~~~ C+~~~~~~ ~~~ ~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. 22,000 0~~~~~~~~~~~~~~~~~ 21,000 2 0,OO00_ 19,000 -400, -300 200 -100 0 +100" DISTANCE ON MICROPHOTOMETER TRACING

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGANTABLE III POSITION AND WAVE NUMBER OF PEAKS IN MICROPHOTOMETER TRACING Rotational Quantum Number Distance Wave Length Wave Number K s _A. A v. 1/cm. 3 - 4.6 4300 23256 4 - 6.2 4294 23286 5 7.8 4289 23317 6 - 9.4 4283 23348 7 -11.0 4278 23378 8 -12.6 4272 23408 9 -14.2 4267 23438 10 -15.8 4261 23469 11 -17.5 4255 23501 12 -19.2 4249 23533 13 -20.8 4244 23563 The relative spectral intensity I(K) at any peak is related to the quantum number K as shown in Equation (6). h C I(K) =( 1 -BK(K+l)e (6) where B = 14.5 Cm-l for CH molecule and hc/k = 1.43. Defining F(K) as F(K) = K2 K(K - 2) 2K - 1 2K - 3 Equation (6) becomes B K(K+l)h kT I(K) = F(K) e (7) F(K) and K(K+1) are tabulated below: 36

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY:OF MICH-IGAN TABLE IV QUANTUM FUNCTIONS K F(K) K(K+1) 3.800 12 4.686 20 5.634 30 6.607 42 7 o 587 56 8.574 75 9.565 90 10.557 110 11.551 132 12.547 156 13.544 182 For a given spectrographic exposure, F(K)/I(K) was obtained and InF(K)/I(Kj was plotted against K(K+l). The slope of this curve was obtained and equated to Bhc/kT, i.e., 14.5 x 145 slope = m = 1 -. T where T = degrees Kelvin in order to obtain a value of the rotational temperature of CR for that particular spectrographic exposure. A typical plot of In[F(K)/I(K] versus K(K+1) is shown in Figure 23. The data for other such plots are tabulated in the Appendix. The plot is not linear at the low quantum numbers but becomes very close to a straight line at the higher quantum numberso For this reason only quantum numbers above 8 were used in determining the slope by least squares. Gaydon and Wolfhard (12)found a similar nonlinearity in determine ing rotational temperatures of OH molecules in low-pressure flames. Computed CH Rotational Temperatures The CH rotational temperature is plotted versus distance through the flame in Figure 24. The maximum temperatures were then plotted versus 37

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN O (O O oO c ____ o o_ 4-) U) o o L.. - 38

m m m z im41 W~~~~~~~~~~~~~~~~~ T cr_~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~J 5~~~~~~~~~~~~~~~~~~~~~~~~~~~~~/o00008 - I'~~~~~~~~~~~~~~~C 6 4000 H 5000 < -~IO"Hg ]' I:5 o ~ ~ ~ ~ ____ __ P.8: 1 4"I'Hg.___ 30 0 0 _ __ __ _ _ _ _ _0~~~~~~~~~~~~~~~~~ Iz r 2000 " to I00'Q)., 0 0.03 0.04 0.05 0.06 0 0.03 0.04 0.05 0 0.02 0.03 0.04 ELEVATION THROUGH FLAME -inch-(arbitrary zero) Z Fig. 24. CH rotational temperature within an irradiated flame

6000 _.. m z J ~~m 5000. -' U.~~~~~~~~~~~~~~~~~~~~~~~~~~~n w 400C i-1I PT=10" Hg -' = -I A 0~~~~ 0 3000 "' - PT=S"Hg r 2000 0.0001,,,01.01 RELATIVE INTENSITY OF RADIATION (I/I*) Fig. 25. Effect of radiation on CH rotational temperature -m J I~~~~~~~~~~~~~~~~ 0 <[,~~~~~~~~~~~~~~~~ Z -d~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 2000 0.0001 0.001 0_.0 0.1 I-~~~~~~~~~~RLTV INTESIT OFRDAIN(/ / ~~~~z

ENGINEERING RESEARCH INSTITUTE *'' UNIVERSITY OF MICHIGAN relative intensity of radiation at several pressures, as shown in Figure 25. The temperature was not plotted as a function of fuel-air ratio because of inability to read the relative spectral intensity, I(K), with sufficient accuracy. This was also true of the data obtained at pressures of 6 and 8 in. Hg at a relative intensity of radiation of 0.0008~ The values of I(K) are particularly difficult to obtain when the intensity of the light emitted in the 4315 A band is low. In Figure 25 it can be seen that the rotational temperature increases with an increase in relative intensity of radiation. This suggests that the CH molecules are more energetic at higher intensities of radiation. This should result in a more rapid rate of reaction with oxygen and perhaps explains the increase in flame speed. At a pressure of 14 in. Hg abs., the CH rotational temperature at the optimum location in the flame was increased 47 per cent by increase in intensity of radiation from 12 curies to 1500 curies. At a pressure of 10 in. Hg abs., the CH rotational temperature is increased by 42 per cent, while at a pressure of 8 in. an increase of 95 per cent was observed for the same increase in radiation. Quantitative significance should not be attached to these percentages due to the limited data. C. SPECTRAL INTENSITIES It is seen from Figures 14, 16, and 19 that the intensities of the three bands are almost symmetrical about the elevation at which the individual maxima occur. This indicates that the maximum rate of reaction occurs midway between the initiation of reaction and the end of reaction, and seems to have no other significance. However, Figures 14 and 19 do show a definite decrease in the intensity with decrease in pressure, indicating a decrease in the quantities of CH, C2, and OH, respectively, with decrease in pressure; and a shift of the maximum with pressure, indicating that at lower pressures the reaction occurs farther away from the screen. Figure 16 indicates a decrease in the intensity of CH with decrease in fuelair ratio, which is to be expected because of the decrease in the quantity of fuel present. The elevation at which the maximum intensity of the CH band occurs does not appear to be much of a function of the fuel-air ratio. It can be seen in Figures 15 and 20 that the maximum intensities due to CH and C2 increase with radiation and also with pressure. Another interesting fact is also illustrated; the slopes of the curves are greater at the higher pressures. This means that the radiation has a greater effect on emission at higher pressures than at lowerpressures, which is to be expected since the greater density of gas at the higher pressures results in a greater probability of beta-ray capture. At 6 in. Hg, the radiation has very little effect on the spectral intensity. The percentage increase in intensity due to CH at 1500 curies over that at 12 curies ranges from about 41

m Z II1 I IoL0.094 I To.0043 I/I.O.0008 0.90 -__ 0.8 315 X315 0.7_ CH-4315 A CH-4315i I~: ~ ~~~~~~~~~~~~~~ 0.6 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ] — I z 0 z C2-5165 - 0.: L 0.3 0.2 C2 -5165 C2-5165 A W - 0.1l r~ A~OH-3063A OH-3063A OH-3063A 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 ELEVATION THROUGH FLAME-inch-(arbitrary zero} 3I Fig. 26. Comparison of emission at 4315, 16 A, and 3063 A within irradiated flames pressure =14 in. Hg, fuel/air = 0.0802 - 0.0883 Z

' 0.07.. m 0)~~~~~~~~~~~~~~~~~~~~~~~~~~~ N G~ 7% 0.0 o m ~o.i.4-..O ~~~~~~~~~~~~~~m 0.06 w ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0 14.-..iHg a. 0 C~~~~~~~~~~~~~H-4315 Al PT= Z (n ~~ ~ ~ ~ ~ ~ ~ ~ ~~ I0.05 ------ I ~~ ~ ~5~~~~~~;;~~~~~~o~~~~3~~:P~~~~~06 X PT=1aH I-Cr Uw. C2__15 _ A, _H: 0.03 I -- 4 m w 0.02 - - -- --—...... O-Cz-5165 0H-504315A P 0 ~~m I, 0.01 c —- - ~-mo~jII I I -r z O RELATIVE INTENSITY OF RADIATION C Fig. 27o Effect of radiation on location of -0maximum emission at 4315 - 1 306 Iii ~~~~~~~~~~~~~~0 w, 0 _ _ _ _ - - - - _ _ _ -_ 0.0001 00100. RELATIVE INTENSITY OF RADIATION 0 1 Fig. 27. Effect of radiation on location of maximum emission at 4515 A, 5i65 A, and. 3063 A z

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN 14 per cent at 6 in. Hg to about 33 per cent at 14 in- Hg, The corresponding increase in intensity due to C2 ranges from about 145 per cent at 8 in. Hg to about 152 per cent at 14 in, Hg. Figure 17 illustrates more clearly than Figure 16 that the quantity of CH present decreases very abruptly with decrease in fuel-air ratio. Figure 18, which is a cross plot of Figure 17, shows the increase in the intensity due to CH with an increase in radiant intensity at various fuel-air ratios. The slopes indicate a greater effect due to radiation at higher fuel-air ratios than at lower ratios. This is to be expected since more fuel molecules or radicals are present to capture beta rays. The per cent increase in spectral intensity at 1500 curies over that at 12 curies ranges from about 57 per cent at a fuel-air ratio of 0.06 to about 68 per cent at a fuel-air ratio of 0.08. In order to compare the data for the three different species, the spectral intensities at 4315 A, 5165 A, and 3063 A were plotted versus elevation through the flat flame at a constant pressure of 14 in. Hg abs. and a nearly constant fuel-air ratio of 0.085 for three relative intensities of radiation (0.094, 0.0043, and 0.0008). These plots are shown in Figure 26. The maxima-in the spectral intensities occur at different elevations for the CH, C2, and OH. The elevation in the flame at which the maxima occurred are plotted versus relative intensity of radiation at various pressures and constant fuel-air ratio in Figure 27. Contrary to expectations, the maxima in the spectral intensities are shifted away from the burner head by the beta radiation. Furthermore, the maximum intensities due to C2 and OH occur at the same elevation, followed by the maximum intensity due to CH at a higher elevation within the flame. This lends support to the theory that carbon monoxide is formed in flames from the reaction C2 + OH e~ CH + CO PART VII. SUMMARY AND CONCLUSIONS The influence of nuclear radiation from Au.198 and Au-199 on propane air flames at low pressure was investigated experimentally. The intensity of light emission by CH, C2, and OH molecules was measured, and the rotational temperature of the CH molecule was computed as a function of distance through the flame. Flame speeds were also measured. The following effects were observed, presumably due to beta rather than gamma radiation 44

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN 1. Rates of flame propagation determined from photographs of bunsen flames were increased up to 50 per cent by nuclear radiation from the 12000-curie source. 2. The intensity of radiation at 4315 A due to CH was increased by 14 per cent at 6 in. Hg abs. and up to 33 per cent at 14 in. Hg abs. by nuclear radiation at a level of 1500 curies. 3. The intensity of radiation at 5165 A due to C2 was increased up to 150 per cent by nuclear radiation at a level of 1500 curies. 4. The intensity of radiation at 3063 A due to OH was not affected significantly by nuclear radiation at levels up to 1500 curies. 5. Increases in the CH-rotational temperature ranging from 50 to 100 per cent were calculated from a limited amount of experimental data obtained at source strengths up to 1500 curies. 6. Under the influence of radiation, the maxima in the emission due to C2 and OH were observed to occur at an earlier stage of the flame than the maximum of emission due to CH. This lends support to the theory that carbon monoxide is produced in flames by the reaction C + OH = CH + CO 2 45

I

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN APPENDIX TABLE V FUNCTIONS FOR CALCULATION OF CH-ROTATIONAL TEMPERATURES K I(K) I(K)/F(K) K(K+1) I/Io = 0 94, F/A = 0.0802 - 0.0883, Pressure = 14 psia Elevation = 0.030 in. 9 0.073 7.74 90 10 0.062 8.99 110 11 o.o43 12.82 132 12' 0.046 12.10 156 13 0.024 22.61 186 T = 11850K Elevation = 0.034 in. 9 0.124 4.55 90 10. 095 5 87 110 11 0.083 6.64 132 12 0.072 7.59 156 13 0o062 8.75 186 T = 30770K Elevation = 0.038 in. 9 0.110 5.13 90 10 0.098 5.69 110 11 0.078 7.07 132 12 0.066 8.28 156 13 0.061 8.90 182 T = 3379~K 47

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN TABLE V (cont.) K I(K) I(K)/F(K) K(K+1) I/Io = 0.094, F/A = 0.0802 - 0.0883, Pressure = 14 psia Elevation= 0.043 in. 9 0.057 9.91 90 10. o46 12.11 110 11 0.039 14.14 132 12 0.028 19.52 156 13 0~028 19.38 182 T = 2669~K I/Io = 0.094, F/A = 0.0802 - 0.0883, Pressure = 10 psia Elevation = 0.043 in. 9 o0.04 10.46 90 10o 0.041 13.59 110 11 0.034 16.22 132 12 0.024 22.78 156 T = 16590K Elevation = 0.047 in. 9 0.076 7.43 90 10 o.o061 9.14 110 11 0.050 11.03 132 12 o.o48 11.39 156 T = 2841~K Elevation = 0.051 in. 9 0.058 9.74 90 10 0.052 10.72 110 11 0.040 13.79 132 12 0.020 27.73 156 T = 24060K 48

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN TABLE V (cont.) K I(K) I(K)/F(K) K(K+1) /Io = 0.94, F/A = 0.0802 - 0.0883, Pressure = 10 psia Elevation = 0.056 in. 9 0.047 12.07 90 10 0.040 13.93 110 11 0.025 22.06 132 12 0.015 36.44 156 T = 1057~K /I = 0.094; F/A = 0.0802 - 0.0883, Pressure = 8 psia Elevation = 0.051 in. 9 0.033 17.11 go 10 0 031 17.98 110 11 o 0.022 24.85 132 12 0.020 27.33 156 T = 23510K Elevation = 0.056 in. 9 0.052 10.86 90 10 0.047 11.86 110 11 0.042 13.13 132 12 0.036 15.18 156 13 0.023 23.59 182 T = 2610~K Elevation = o.o60 in. 9 0.062 9.11 90 10 0.059 9.45 110 11 0.031 17.79 132 12 0.024 22.78 156 13 0.024 22.61 186 T = 19940K 49

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN TABLE V (cont.) K I(K) I(K)/F(K) K(K+1) I/Io = 0.0043, F/A = o.o834 - 0.0852, Pressure = 14 psia Elevation = 0.030 in. 9 0.070 8.07 90 10 0.052 10.7 110 11 0.041 13.4 132 12 0.038 14.4 156 T = 2103~K Elevation = 0.034 in. 9 0.10 5.65 90 10 0.073 7.63 110 11 0.059 9.35 132 12 0.059 9.26 156 T = 24480K I/Io = 0.0043, F/A = 0.0834 - 0.0852, Pressure = 10 psia Elevation = 0.038 in. 9 0.032 17.6 90 10 0.025 22.3 110 11 0.019 29.0 132 T = 1714~K Elevation = 0.043 in. 9 0.036 15.7 90 10 0.028 19.9 110 11 0.024 23.0 132 12 0.020 27.3 156 T = 22300K 50

ENGINEERING RESEARCH- INSTITUTE' *..iUNIVERSITY.. OF- MICHIGAN TABLE V (cont.) K I(K) I(K)/F(K) K(K+1) I/Io = 0.0043, F/A = 0.0o834 - 0.0852, Pressure = 10 psia Elevation = 0.047 in. 9 0.024 23.5 90 10 0.026 21.4 110 11 0.019 29.0 132 12 0.014 39.0 156 T = 21553~K I/Io = 0.0008, F/A = 0.0829 - 0.0874, Pressure = 14 psia Elevation = 0.038 in. 9 0.040 14.1 90 10 0.030 18.6 110 11 0.026 21.2 132 12 0.021 26.0 156 T = 1957~K Elevation = 0.043 in. 9 0.051 11.1 90 10 0.037 15.1 110 11 0.032 17.2 132 12 0.028 19.5 156 T = 2400~K Elevation = 0.047 in. 9 0.042 13.4 90 10 0.030 18.6 110 11 0.026 21.2 132 12 0,28 19.5 156 T = 1000~K 51

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN TABLE V (concl.) K I(K) I(K)/F(K) K(K+1) I/Io = 0.0008, F/A = 0.0829 - 0.0874, Pressure = 10 psia Elevation = 0.030 in. 9 0.10 5.6 90 10 o0.0o8 7.0 110 11 o.o65 8.5 132 12 0.060 9.1 156 T = 2250~K I/Io = 0.0008, F/A = 0.0829 - 0.0874, Pressure = 8 psia Elevation = 0:030 in. 9 0.026 21.7 90 10 0.021 26.5 110 11 0.015 36.8 132 12 0.011 49.7 156 T = 1450O~K 52

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN BIBLIOGRAPHY 1. Cullen, R. E*, and Gluckstein, M. E., "Effect of Atomic Radiation on the Combustion of Hydrocarbon Air Mixtures", Fifth Symposium (international) on Combustion, Reinhold Publishing Corp., New York, 1955, P. 569. 2. Morrison, R. B., Cullen, R. E., and Weir, A., "Utilization of Gross Fission Products; Performance of Combustion Engines under the Influence of Radiation - Jet Engines", A.E.C. Contract No. AT(11-1)162, Progress Report No. 2, January 31, 1952. 3. Morrison, R. B., Cullen, R. E., and Weir, A,,'"Utilization of Gross Fission Products; Performance of Combustion -Engines under the Influence of Radiation - Jet Engines", A.E.C. Contract No. AT(111=)162, Progress Report No. 3, June 30, 1952. 4. Weir, A., Jr., Ind. Eng. Chem., 45, 1637 (1953). 5 Weir, A., Jr., and Morrison, R. B., "Basic Research on Ceramic Combustion Chambers", Engineering Research Institute Report No. 2054-3-F, University of Michigan, September, 1954. 6. Supplement No. 3 to NBS Circular 499, "Nuclear Data", Department of Commerce, National Bureau of Standards, Washington, D. C., June, 1952. 7. Lew-is, W. B.., "Isotope Production: How to Choose Irradiation Time", Nucleonics, 12, 30-33 (1954). 8. Lewis, W. B., Private Communication. 9. Cullen, R. E., "The Effect of Pressure on the Propagation Rate of Bunsen Flames in Propane-Air and Ethylene-Air Mixtures", U.S.A.F. Contract No. W33-038-ac-21100, December, 1950. 10. Cullen, R. E., Trans. A.S.M.E., 75, 43 (1953). 11. Pearse, R. W. B., and Gaydon, A. G., "Identification of Molecular Spectra", 2nd Edition, John Wiley and Sons, Inc., New York, 1950. 12. Gaydon, A. G., and Wolfhard., H. G., "Flames, Their Structure, Radiation, and Temperature", Chapman and Hall, Ltd.., London, 19535. 53

UNIVERSITY OF MICHIGAN 3 90111111115 02828 4050 3 9015 02828 4050