ENGINEERING RESEARCH INSTITUTE THE UNIVERSITY OF MICHIGAN ANN ARBOR Annual Report EFFECT OF GAMMA RADIATION ON VARIOUS PETROCHEMICAL REACTIONS August, 1955 to July, 1956 P _ roject Supervisors Joseph J. Martin Kenneth F. Gordon Research Assistants Frank DeMaria William V. Dicke Richard Schwing Wayne Kuhn Project 24t20 STANDARD OIL COMPANY (INDIANA) Noveiniber 1956A'RT

The University of Michigan * Engineering Research Institute TABLE OF CONTENTS Page LIST OF TABLES LIST OF FIGURES iv ABSTRACT vi OBJECTIVE ii I. INTRODUCT I ON 1 II. RADIATION FACILITIES 1 III. EXPERIMENTAL 4 A. Batch Experiments at Room Temperature 4 1. Equipment and Procedure 4 2. Reactions Studied and Results 7 B. Batch Experiments at Elevated Temperature 10 1. Equipment and Procedure 10 2. Polymerization of Pentene-l 13 3. Constant-Volume Cracking of Cetane 15 4. Cracking of Cetane at Total Reflux 15 C. Continuous Thermal Cracking of Cetane 18 1. Equipment and Procedure 18 2. Results 20 5. Thermocouple Calibration 24 D. Catalytic Cracking of Cetane 26 1. Equipment and Procedure 26 2. Results 27 IV. INTERPRETATION OF RESULTS ALND CONCLUSIONS 29 A. Gamma-Radiation Effects at Room Temperature 29 B. Batch Experiments at Elevated Temperature 29 C. Continuous Thermal Cracking of Cetane 52 ]D. Catalytic Cracking of Cetane 55 APPENDIX 34

The University of Michigan * Engineering Research Institute LI3T OF TABLES No. Page I. Irradiation of Methyl-Cyclopentane-Butylene Mixtures 8 II. Irradiation of 2-Methyl-Pentane-Pentene-1 Mixtures 8 III Irradiation of n-Heptane-Pentene-1 Mixtures 9 IV. Irradiation of 2-Methyl-Pentane-Pentene-1 Mixtures 11 V. Polymerization of Pentene-l 1 VI. Cracking of Cetane-Summary of Results 16 VII. Continuous Pyrolytic Cracking of Cetane-Summary of Results 21 VIII Thermal Cracking of Cetane Gas Analysis 22 IX. Thermal.-Cracking Temperature Corrections 25 X. Catalytic Cracking of Cetane.- 28

The University of Michigan * Engineering Research Institute LIST OF FIGURES No. Page 1. Cobalt rod for 10-kilocurie source. 2 2. The 10-kilocurie cobalt-60 gamma-ray source. 2 3. Cutaway view of lO-kilocurie gamma-irradiation facility. 3 4. Dose rates on planes normal to source. 5 5. Glass vials. 5 6. Filling apparatus for vials. 6 7. Photo of reactor. 12 8. Constant-pressure reactor. 12 9. Constant-volume reactor. 13 10. Pentene-1, pressure vs temperature, Run No. 158683. 14 11. Pentene-1, heating cycle, Run No. 158683. 14 12. Pressure vs time for cracking of cetane, Run No. 158756. 17 13. Pressure vs time for cracking of cetane, Run No. 158775. 17 14. Continuous flow reactor. 19 15. Cetane present in product vs temperature. 23 16. Liquid-product distribution vs temperature. 23 17. GaseoLus-product distribution vs temperature. 24 18. Error in temperature measurement. 25 19. Catalytic cracking unit flow diagram 26 iv

The University of Michigan * Engineering Research Institute LIST OF FIGUIRES (Concluded) No. Page 20. Catalyst arrangement in reactor with spacer. 27 21. Catalyst arrangement in reactor without spacer 27 22. Slope of linear pressure increase vs reciprocal temperature for cracking of cetane. 0 25. End point of linear increase in pressure vs reciprocal temperature for cracking of cetane. 1 24. Total pressure after one hour at reaction temperature vs reciprocal temperature for cracking of cetane. 31 25. Thermal cracking, approximately 1-hr contact time, moles of gas per mole of feed vs temperature, 0C. 5 26. Integration for radiation intensity. 27. Vapor pressure vs temperature for compounds studied. A.P.I. Project 44. 7 28. Specific gravity of cetane vs temperature, ~C. A.P.I. Project 41. 8 v

The University of Michigan * Engineering Research Institute ABSTRACT Gamma radiation emitted from a nominal ten-kilocurie cobalt-60 source has been used to investigate its effect in initiating some petrochemical reactions. Approximately a half percent or less conversion was observed for the following mixtures sealed in glass vials and with total doses from 5 to 30 megarep: methyl-cyclopentane, butylene 2-methyl-pentane, pentene-1 heptane, pentene-1 pentene-l. methanol Batch experiments at temperatures from 150~ to 300~C on pentene-l showed that a certain degree of polymerization occurred with and without radiation. The extent of the possible effect of radiation was not determined. Batch experiments on the thermal cracking of cetane at temperatures varying from 287~ to 547~C indicated that gamma radiation inhibits gas and coke formation. Reflux experiments gave further evidence that gamma radiation inhibits gas formation. Additional work in which cetane was thermally cracked in a continuous system did not show as much difference between irradiated and nonirradiated runs as in the batch and reflux work; however, it appears that consistently more gas was made in the nonirradiated runs. Catalytic cracking studies of cetane are in progress. Preliminary results based on the moles of gas formed indicate that the nonirradiated runs produced more moles of gas per mole of feed than the irradiated runs. A more complete analysis of this work will have to await the completion of the chemical analysis of the products as well as additional experimental work now in progress. vi

The University of Michigan * Engineering Research Institute OBJECTIVE The aim of this work is to explore the effects of gamma radiation on a number of chemical reactions which may be of interest to the petroleum and petrochemical industries. It is believed that these exploratory studies will lead to a better understanding of the mechanism of radiation-induced reactions. Also, such studies should permit generalizations of radiation effects on many different chemical reactions.

The University of Michigan * Engineering Research Institute I. INTRODUCTION A study of the effect of gamma radiation on various chemical reactions was undertaken for the Standard Oil Company (Indiana) under Project No. 2420 in the Engineering Research Institute of The University of Michigan. The present report embodies the work from August, 1955, to July, 1956, inclusive. The research program consisted of two distinct phases. In the first, various reactions were screened in batch experiments to obtain qualitative information on the effect of gamma radiation. The systems investigated were: a) Methyl-cyclopentane, butylene b) 2-Methyl-pentane, pentene-l c) Heptane, pentene-l d) Pentene-l, methanol e) Pentene-l (up to 200'C) f) Cetane (up to 485~C) The second phase of the work, representing the major part of the research effort, consisted of a more detailed study of the thermal and catalytic cracking of cetane. II. RADIATION FACILITIES A cobalt-60 source nominally rated at 10 kilocuries was used in the radiation experiments. It consists of 100 cobalt rods enclosed in aluminum sheaths as shown in Fig. 1. The rods are arranged in two concentric rows at the periphery of a cylindrical rack as shown in Fig. 2. Details on the construction and activation of the cobalt-60 source are to be found in Fission Products Laboratory Progress Report No. 4 (COO-124),1 for the Atomic Energy Commission. The source is housed in a radiation cave adjacent to the Fission Products Laboratory, a cutaway view of which is presented in Fig. 3. The sixteen-foot well shown there is filled with water inl which the radi ation source is submerged in order to allow access to the cave. The source in the raised position is protected by a wire-mesh cage of approximately the same dimensions 1

0.405" 0.269" 5 -0.250" SECTION IB-IB 0~~~~~~~~~~~~~~~~~~~~~~~ 3SHI18 ALU MINUM PIPE 1/8IPS 0,: ~':.:' 10-1/8" ALCAN 25 WELDING 0 ROD _ _ _ SECTION Af-A Fig. 1. Cobalt rod. for lO-kilocurie source. Fig. 2. The lO-.kilocurie cobalt..60 gamma-ray ore

The University of Michigan * Engineering Research Institute z 03:~~~~~~~~~~~~~~~~~~~~~~~~~~~~: - 32~~::::::C,) Om~~~~~~~~~~~~~~~~~c r-d ct3 CH J Wt.,,.,;.................. O~~~~~~~~~~~~~~~~~~~~~~- rx~~~~~~~~~~~~~~~~~~~~ (L): % ' Y.s -- 'n wi.............;. / /. s g, X iffi o.S.....;;, S. v....... o W F fL.,4............. Z~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. z~~~~~~~~~~~~~~~~:: ' 0= cli b0 H _- 3~

The University of Michigan * Engineering Research Institute as the cylindrical rack. For the maintenance and routine operation of the source cave, reference is made to Fission Products Laboratory Progress Reports No. 5 (C00oo-196)2 and No. 6 (C00oo-198)3 for the AEC. The reaction vessels to be irradiated were placed as close as feasible around the protective cage in order to take advantage of the highest possible radiation rates. The position of the reactor with respect to the source was carefully determined during each experiment, Wherever possible, the centerline of the reactor was adjusted so that the midplane of the reactor contents coincided with the centerplane of the source rods. For the 15-cc glass vials used in this work it was possible to obtain large radiation rates by utilizing the center core of the source. Prolonged irradiations of batch experiments in glass vials were conducted intermittently because of the need for allowing periodic access to the cave, The net irradiation time of such experiments was computed with the aid of a running log book kept at the cave entrance. In general, all experiments at higher temperature than ambient were conducted by maintaining the source in its raised position for the whole duration of the run. The actual strength of the source is only approximately 30 percent of its l0-kilocurie ratingo A thorough description of the radiation field of the source and the procedure used in its determination was made by Lewis and Martini in March, 1953 Figure 4 gives dose-rate curves in kilorep/hr versus distance from centerline of source along planes normal to the source. The values presented there were calculated from the data of Lewis and Martin, allowing for the decay of cobalt-60 up to March 31, 1956, the half life being 5.3 years. The dose rate in each experiment was calculated with the aid of this graph. A sample calculation for dose rates in a heated metal reactor is given in the Appendix. It should be noted that the dose rate calculated in this manner for the metal reactor agrees with a ferrous sulfate dosimetric determination recently performed within the limits of error reported by J. Weiss.5 IIIo EXPERIMENTAL A. BATCH EXPERIMENTS AT ROOM TEMPERATURE 1. Equipment and Procedureo.-Glass vials of 15- and 100-cc capacity made from standard heavyvwall pyrex tubing were used in the screening experiments at room temperature. A diagram of these vials with their dimensions is given in Figo 5. Considerable precaution was takenin filling the vials to be irradia

The University of Michigan Engineering Research Institute 120 Source rods 110 T.-~~!1 ~ICI0~ ~Midplane - 1IT of source 100 90 a::: 80 a. Midplane Y glL ~~~~~~6" w6O.: ~47" ~ 50:, I d o::" 40 70 30 00 0 -I8 8 2 4 6 8 10 12 14 INCHES FROM CENTER LINE OF SOURCE Fig. 4- Dose rates on planes normal to source. ~OHEAVY WALL 18-mmI PYREX TUBE It HEAVY WALL 26 -mm PYREX TUBE'I I 15,cc VIAL In50 —.,IA.o c3go N INC~~~~~HEAVYMCETRLIEO SUC WL18-rmm PRXTUBE ~ HEAVY ________ mm PYREX TUBE 15-cc VIAL IO0-cc VIAL SEALED VIAL Fig~ 5o Glass v~also 5i

The University of Michigan ~ Engineering Research Institute diated in order to avoid oxygen contamination. The equipment used is shown schematically in Fig. 6. It consisted of a vacuum manifold, a gas-measuring apparatus, and a vacuum pump. VENT ___ 1 — ' -_TO MERCURY CATCH POT IN BACK OF BOARD TO VACUUM PUMP TO GAS BOTTLE 7 | | MANOMETER 7ILLING ~V ~~~~~ENT ( -PORTS MEASURING BULB_ FLAME SHEILD LEVELING BOTTLE WATER VIAL BATH PROTECTIVE VESSEL THERMOMETER Figo 6. Filling apparatus for vials. The vacuumn manifold, made of pyrex glass, provided the connection bet ween 'the vacuum pump and the vial. 'to be evacuatedo It was equipped with an open mercury manometer, two gas fill:ing ports, a:nd a vent. The vaculu pump was a 'Duo-seal, Welch PLunp mounted, on a portable ata3nd and equipped with a closed mercury manometer, a McLeod, gauge, and a cold trap. The purp wa. eorrnected to 'the -vac~uun manifold by means of rubber tubing. The gas -fill..iig apparatus coi.nis.ted of a one -liter calibrated sphericeal fla.sk immnersed. in a water bath. The flask was connected to a leveling bottle which permitted the filling of the flask at atmospheric pressure and 'tlhe lis.charge of a. measured volume of gas. in the vial. The filling operationa was>)- made by connecting a vial as shown in Fig. 6 and evaculating it to a pressure of 100 microns or less. A heater was placed around the vial to bring the temperature of the viaL up to about 200~Co The vial ~was then repeatedly purged with nitrogen introduced from one of the lefth:andside filling ports. After the system had been flushed with nitrogen at leas t three times, tle heater wa~s removed and. the vial allowed to cool to room ter-mperatlure~ It wa.s further cooled bvy immrsing it in a chloroform-carbontetrachloride.dry-ie mixture~ The gas to be charged wras introduced into the measuring bulb and, wh n the tempelrature of the vial reached equilibriumn the gas wras introduced into the vial Coiodansatio n occrred very rapidly if the

The University of Michigan * Engineering Research Institute gas was the first reagent charged to the system. This gas-filling operation was repeated as many times as necessary in order to transfer the desired weight of gas to the vial. The gas-measuring apparatus was then disconnected and replaced with a graduated burette to deliver the desired amount of liquid reagent in the vial. The vial was sealed with an oxygen flame below the tee connection, making certain that the vial was still under vacuum. All connections were made with neoprene tubing to minimize gas permeability. The filling of components which were both liquids at room temperature was more conveniently accomplished by introducing the liquids into the vials by means of a syringe piercing through a standard serum bottle stopper inserted into the side arm. 2. Reactions Studied and Results. — a. Methyl-Cyclopentane, Butylene. Five glass vials were filled and irradiated. The reactants, furnished by the Standard Oil Company (Indiana), were especially purified in their laboratories to reduce impurities to a minimum. Conditions of the experiments, proportion of reactants, and radiation dosage are summarized in Table I. No visible change in the appearance of the reactants after irradiation was noted. b. 2-Methyl-Pentane, Pentene-1. A total of eleven 15icc vials were charged with varying proportions of 2-methyl-pentane and of pentene-1. The pentene-1 was Phillips, 99 mol percent pure grade, and was purified by passing through a column packed with Davison Silica gel. The column used was 3/4 inch in diameter and 5 feet long. The purified pentene-l was stored under nitrogen atmospheres and always handled in the presence of nitrogen. The 2-methyl-pentane was furnished already purified by the Standard Oil Company (Indiana). The experimental conditions are summarized in Table II. No visible change in the appearance of the reactants was noted. c. Heptane, Pentene-l. A summary of the conditions for the seven experiments made for the system heptane, pentene-l1 is given in Table III. Heptane was furnished by the Standard Oil Company (Indiana) and was used without further treatment. Pentene-1 received a purification treatment as described in (b). Here also no visible change in the appearance of the reactants was noted.

TABLE I IRRADIATION OF METRYL-CYCLOPENTANE-BUTYLERE MIXTURES Distance Bath Boom Butylene Mol. St. of o Vial Time o Approx. Total Run No. Date Temp. Temp. cc MOP Ratio* Liquid from Exposure Radiation Dose Comments 'F 'F Atm. cc Butylene/ in Vial Source hr Rate mep Press. MCP (in.) (in.) krep/hr 1-158653 11/4 67 72 3220 15.8 1.1 4 4 112.4 41.6 4.7 100-cc vial. 2-158656 11/10 63 74 5220 15.8 1.1 4 4 527.5 41.6 15.7 100-cc vial. 3-158657 11/11 64 74 5220 15.8 1.1 3.5 4 670.8 41.5 28.1 100-cc vial. Some difficulty in handling may have caused some 02 contamInation. Q 4-158657 11/15 66 72 5220 7.8 2.0 5.5 4 579 41.5 24.1 100-cc vial. Volume of MOP only approximate. 17-158668 12/9 -- -- -- 5 -- 1.75 87.8 180 15.8 15-cc vial. Reight of liquid above center wqll, 1.5 in. *Mol. ratio computed on basis of compressibility factor for butylene of 0.97. Sp. Or. of methyl cyclopentane.744. 0C TABLE II IRRADIATION OF 2-METHYL-PENTANE-PENTERE-l MIXTURES Vol. Ratio St. of Ht. of Bottom Time of Approx. Total Run RN. Oats Vol. 2-Methyl Mol. Liquid of Liquid Exosue Radiation Charged Pentane- Ratio in Vial Above Center hr Rate moeC Pentene-1 (in.) Well Floor krep/hr 1-158658 11/19 15 4.0 3.3 4.0 1.0 39.5 180 7.1 Loaded with burette instead of syringe. 2-158659 11/25 5.0 4.0 5.5 1.75 1.0 118.6 180 21.5 Contaminated with stopcock grease. Bubbles appeared. 3-158659 11/25 5.0 4.0 3.3 1.75 1.0 118.6 180 21.5 4-158660 11/25 5.0 4.0 3.3 1.75 1.0 87.6 180 15.8 0 5-158662 11/26 5.0 9.0 7.5 1.75 1.0 87.6 180 15.8 6-158662 11/26 5.0 9.0 7.5 1.75 1.0 63.8 180 11.5 7-158665 11/26 5.0 9.0 7.5 1.75 1.0 63.8 180 11.5 8-158663 11/26 5.0 4.0 5.5 1.75 1.0 28.6 180 5.1 9-158664 11/28 5.0 4.0 5.5 1.75 1.0 28.6 180 5.1 18-158668 12/9 5.0 100% Pentene-1 1.75 1.1 87.8 180 15.8 19-158670 12/10 5.0 100% 2-Methyl Pentane 1.75 1.1 87.8 180 15.8

rTABLE III 'l IRRADIATION OF n-HEPTATE -PENTENE-1 MIXTURES 0 Ht. of 3Ht. of Bottom Vol. Vol. Ratio Mol. Liquid of Liquid Time of Approximate Total Mel, Eiqufdof Liquid Run No. Date Charged Heptane Ratio in Vial Above Center Exposure Radiation Dose Ra~tio in Vial AoeCne:= cc Pentene1 (in.) Well Fr. hr Rate, krep/hr mrep ' in,) Well Flr. l0-158664 12/5 5.0 4.0 3.0 1.75 1.5 99.5 180 17o 9 1 1-158665 12/5 5.0 4.0 350 1.75 1.5 60.75 180 10o9 12-158665 12/5 5.0 9.0o 6,75 1.75 1.1 29~37 180 13-158665 12/7 5.0 9.0 6.75 1.75 1.1 60.75 180 10.9 14-158666 12/7 5.o0 90 6.75 1.75 1.2 99.5 180 17.9 15-158667 12/7 5.0 9.0 6.75 1.75 1.0 29.37 180 5.25 16-158667 12/7 5.0 100% n-heptane 1.75 1.5 87.8 180 15.8

The University of Michigan ~ Engineering Research Institute d. Pentene-l, Methanol. Nine vials were prepared with different concentrations of pentene-l, which was in general maintained in excess to suppress a possible methanolmethanol reaction. To verify such a possibility, one vial with pure methanol was also irradiated. The run conditions for each experiment are summarized in Table IV. Pentene-l was silica gel purified while the methanol was purified over calcium hydride and filtered in an atmosphere of nitrogen. e. Cetane. Two 15-cc vials filled with 5 ml of cetane each were also irradiated (Exp. Nos. 20.158670 and 21-158671). Both were exposed to gamma radiation for a total of 77.2 hours, corresponding to an approximate dose of 14.8 megarep. No apparent change in the appearance of the cetane was noted. The cetane used was supplied by the Standard Oil Company (Indiana). B. BATCH EXPERIMENTS AT ELEVATED TEMPERATURE 1. Equipment and Procedure, —Stainless-steel reactors of standard design made from a 13-inch-long piece of 1-inch stainless-steel heavy pipe were used for all the experiments that follow. Six such reactors were supplied by the Standard Oil Company (Indiana). The construction is as shown in Fig. 7. In order to find a simple reliable procedure to obtain preliminary data, two different types of experiments were designed and tested. One type was operated with a constant-pressure system and thle other with a constantvolume system. The constant-pressure experiments were carried out with the equipment shown in Fig. 8. Here a given amount of reactant was heated to a given temperature. The back-pressure regulator was preset to a pressure slightly above the vapor pressure of cetane at that temperature. The noncondensable gases formed in the reaction increased the pressure of the system and escaped through the valve to be collected in the gas bottle. The rate of gas production could be measured by the amount of water displaced in a given time. Although the constant-pressure experiments presented fewer problems in the control of the reaction temperature, they were found to be more complicated and certainly more difficult to run in the source room and were abandoned in favor of the constant volume experiments. The constant-volume experiments were conducted in the apparatus shown in Fig. 90 This consisted of a pressure vessel described before, equipped with a pressure gage, surrounded by an electric heater. The vessel was charged by evacuating it and then injecting a given amount of reactant through the bottom part of the vessel by means of a syringe. The reactor was heated to a given temperature, and the rate of increase in pressure with time was recorded. 10

TABLE IV IRRADIATION OF 2-METfHYL-PENITANE —PT-ENTENE-1 MIXTUJRES -I.......................................... or" Bott...... Vol. Vol. Ratio Ht, of of Bottom Time of Approximate Total o Pentene-I of Liquid Run No. Chnarged Pentene-l Ml Liquid Exposure Radiation Dose Ratno Above Center cc to Methanol Ratio in Vial e r hr Rate, krep/hr mrep Well Fir, 22-158677 5.0 100% Methanol 175 1.1 81.9 180 14.8 23-158677 5.0 0,47 1.75 1.1 81.9 180 14.8 H 26-158679 5o0 3 1.41 1.75 1.1 81.9 180 14.8 29-15868o 5.0 19 8.95 1.75 1.1 81.9 180 14.8 24-158678 5.0 1 0.47 1.75 1.1 53.55 180 9.6 ' 25-158678 5.0 1 0.47 175 1.1 27.85 180 27-158679 5.0 3.41 1.75 1.1 53.55- 180 9.6, 28-158680 5.o 3 1.41 1.75 1.1 27.85 180 5.O 30-158681 5.0 19 8.95 1.75 1.1 53-55 180 9.6 31-158681 5.0 19 8.95 1.75 1.1 27.85 180.0

The University of Michigan ~ Engineering Research Institute Fig. 7. Photo of reactor. -T.C. 2 WATER OUT 50 CLOSE TURNS 1/4" I.D. COPPER TUBE NITROGEN DOME LOADED B.P.R. WATER IN 1000- WATT o HEATER T.C. I Fig. 8. Constant-pressure reactor. L* <:I -

The University of Michigan * Engineering Research Institute PRESSURE GAUGE RUPTURE DISK NITROGEN VENT 4 VACUUM 1,,, e-PORT POWER LEAD SKIN T.C. TO ~~~~RECORDER 1000I-WATT HEATER FILL PORT POWER LEAD TC. TO RECORDER Fig. 9. Constant-volume reactor. 2. Polymerization of Pentene-1. —Some preliminary work at constant volume was done by treating pentene -1 at temperatures varying from 1500 to 300'C. The results of six experiments are reported in Table V. The pressure-temperature curve and heating cycle for a typical run are shown in Fig. 10 and Fig. 11, respectively. The total final pressure observed in these runs is considerably below the vapor pressure of pentene-l at corresponding temperatures, indicating probable polymerization. TABLE V POLYMERIZATION OF PENTENE-1 Total Final Dose Refractive Run Temp. Heating Color Date 0C Period Pressure Rate Index Product mm psig krep/hr N l 158683 1/19 150 95 277 -- 1.3701 colorless 158752 1/30 200 65 405 -- 1 3759 yellow 158753 2/1 300 45 750 -- 153783 light brown 158758 2/3 200 75 380 -- 1 3790 brown 158762 2/7 200 118 36C 1.3762 light yellow 158765 2/8 200 60 290 42 1.3720 colorless Note: Refractive index of pentene-l ND25 1.3699 - 1'3709 - the variation is due to the poor refractometer used.......... 13..

The University of Michigan * Engineering Research Institute 350 / 300 0/ 250 / a' 20C a 150 / w O! 1 2/ 100 0 Heating period 0 Cooling period -Vapor press. data-Nat. Gas. Sup. Men Ass. ~~~~50 ~ ~ --- Vapor press. data extrpltd. Pressure gauge to crit. cond. not accurate up 0 to this point 20 40 60 80 100 120 140 160 180 200 TEM P. OC Fig. 10. Pentene-1, pressure vs temperature, Run No. 158683. j. 240 0. 00 0 o2 20 40 60 80 I00 120 140 160 180 TIME, MINUTES Fig. 11. Pentene-1, heating cycle, Run No. 158683. 14

The University of Michigan * Engineering Research Institute The results obtained in a single run (158765) under gamma radiation are not considered sufficient to warrant any conclusion. No gas was made in any of -the experiments and the inspection of the products was limited to refractive index measurements which revealed little reproducibility in runs conducted. under the same conditions. 3. Constant-Volume Cracking of Cetane.-A total of 17 runs were made in order to determine the effect of gamma radiation on the thermal cracking of cetane at temperatures ranging from 400~ to 475~C. Run conditions and results are summarized in Table VI. The first seven runs were made by evacuating the constant-volume reactor shown in Fig. 9 and charging 25 cc of cetane by means of a 30-cc syringe connected with rubber tubing to the bottom inlet of the reactor.. The reactor was then rapidly heated to a given temperature which was maintained constant by manual control for approximately one hour. Pressure readings were taken at frequent intervals throughout the experiment. A typical pressure-time curve for Run No. 158756 is shown in Fig. 12. Considerable difficulty was encountered in trying to approach the reaction temperature in a uniform and reproducible manner, especially for comparable runs outside and inside the gamma-source room, due to the appreciable differences in ambient temperature of the two sites. Recourse was made to automatic control of the temperature by using an especially prepared double-wound heater~ The inner winding of the heater was controlled automatically, the outer being kept at a constant current input by manual adjustments. In addition, the experimental technique was modified by heating the reactor to the desired constant temperature, then evacuating it and introducing 25 cc of cetaneo The temperature, after a sharp fall, would increase rapidly, becoming constant after about 15 minutes from the time of injection of the cetane. The length of each experiment was carefully timed from the start of the injection for the duration of exactly 60 minutes. The last nine experiments presented in Table VI were carried out in this manner. A typical pressure-time curve for Run No. 158775 is presented in Fig. 13. The pressure-time curves for both types of experiments are quite similar in that a linear increase in pressure with time is observed in the early part of the experiments and then gradually slopes off toward the end of the run. The slope of the linear pressure increase and the point where the pressure increase begins to depart from linearity are included in Table VI. 4. Cracking of Cetane at Total Reflux, —The large number of variables made control and reprodulcibility difficult in the present apparatus preventing an accurate comparison between irradiated and nonirradiated runs. Therefore, a simple fouling point experiment was made.

TABLE VI [ORA0C1~G OE~ ~~COF `!ETAJ\- WtIAFY- OF RESULTS 0~~~~~~~~~~~~~~~ H H U -H o 'H_1 rd d c o o - 0 H P1 It, P P1 0 P1I 0i 4 -P1 ri0 -0 0L I to 0~~~00 0 crv, 0 0~~~~~~~~~~~~~H 1 0 r~ HI ri -P 0& C IP, 0 % I 0 0 — i 0 P1P1 158755 2/3 47o 7ye 6~ -10-85 0 15.0 brown sol id. 6 970 New bomb us e -a 18-757 2/3 4`5 75 64 1500 42 15.6 dark brown liquid 15.9 122 aN158760 2/4 47'7 52 62 140 42 1.53 akbon iud2. 1522 158764 2/7 400 ~~~~~~~~~~117 47 15 0 17.0 dark yellow solid l.' 7 158766 2/8 400 5-0 61 90 42 17.0 white s ol id 2.1 i58767 2/9 455 55 6o 540 42 15.4 yellow solid 8.6 575f: 158768 2/9 485 52 10 1875 0 10.2 dark brown little liquid 47.i,5 1750 Run interruptedpe. sure increase t r 158771 2/10 455 60 500 0. yAo sld4. 5 New bomb used, 158772 2/11 456 6o 380 0 12.6 dark yell-ow solid 6.0 267 158775 2/15 455 60 35500 10.0 dark yellow solid 6.o 285 18775 2/15 457 60 420, 42 1.1.6 light brown so-lid 7.5 280 158776 2/15 470 60rj 1815 0 __11. 6 dark brown solid 58.0 965 1587 77 2/118 470 - 19 1250 42 14.2 brown solid - 915 Run interruptedshr circuitlu occurred 15878 2/18 4355 - 60 5538 42 — 15.6 yellow solid 7.4 215 158779 2/20 470 60) 141 5 42 15.0 dark brown partly liquid 42.0 765 158781 2/22 470 -- 60 1925 0 10.8 dark brown solid 94.0 1000

The University of Michigan * Engineering Research Institute 1200 1100 C-nEnd of run 1000 800 700 a. U / Constant temp.: 600 reached here -- 500 a. 400 300 200 Cetane charged here at room t00 temperature 0 20 40 60 80 o00 120 140 160 180 TIME-MINUTES Fig. 12. Pressure vs time for cracking of cetane, Run No. 158756. 480 440 400 J'/. -End of run 360 320 280 a. I O 240 C0 )LJ Of 200t Constant temp. 160 reached here 120 80 17

The University of Michigan ~ Engineering Research Institute - -ahisoinvczld. c- of cetl ne were charged to a 200ce three-neck, rourdbottozi f t1a2-htk eqtuipperd. with ref lux cond.enser and a thermocouple immersed in thle liqulid Th11e refliL condenser was connected to a gas bottle and all joints were seali ed. with iDe3Kbaotirsky cement to avoid leakage. The cetane was then heated to boiling (2870C) anrd refluxed for one hour. Fifty cc of gas were liberated in this experiment when no radiation was used. The same experiment was conducted simrl.ltaneously under an approx:mate radiation rate of 80 krep per hour3; no gas wa&s liberated here. Both liquid samples were very slightly discolored. These res]lt;s welre closely reprod3uced by repeating a set of comparison runs on three differen.t days. The collected gas from the nonirradiated reflux reaction was analyzed waith a B3urreil gas analyzer and found to have the following composition: Hydrogen 63.5 mole percent Paracff ins computed as: Methane 14.1 mole percent Ethane 22.4 mole percent No olefins, carbon monoxide, or carbon dioxide were detected. Because of the presence of air over the cetane at the time of charging, the hydrogen and paraffins constituted onlJy about 4 percent of the total gas sample. On the basis of the above result it appears that simple thermal crckieng of cetane tends to split off hydrogen or one or two carbon atoms and theidr ap-:ociated hydrogen. If one assumes that one mole of gas is formed per mole of cetane decomposed, the corresponding convlersion or degree of cracking is less thean ornetenth of a percent dur:ing one hour of refluxing at 2870C, In the radi.ation cracking runs it appears that the cetane molecule is broken near thi;:e centrte of the chain of carbon atomns, which would account for no light gases being liberated, It may b.e noted at t-his point thLat the tenmperature in the reflux experiments shou.ildd have been strictly the same for bothl irradiated and nonirradiated runas because t;he e)periments were conducted at the boiling polint of ceta-tne tunder p&ractically the same atmospheric pressure ozl any given day. An in sPectiion of tlhe re ecords shows, however, that while the temperature o the simple thermal rim.s was 287~C (boiling point of cetane), the temperature of the irradiated rims was, at all times slightly higher. At the time no importance was given to such an anormaly in the temperature recording which was to influence later experiments as discussed in the following pages. C. CO]NTINiUOU, S rI.ERMAL, C(RA.(C.EKING OF CETAI 1. EquLipment and Procedureo —The various components of thle flow system used were supplied by the Standard Oil Company (Indiana). The equipment consisted of a feed tank connected to a "Ruska" positive displacement pump cap18

The University of Michigan * Engineering Research Institute able of delivering a minimum of 0.016 cc/min to a maximum of 3 cc/min. The transfer line from the pump led to the reactor, which is the standard-type vessel used in all other high-temperature experiments (refer to Fig. 7). The reactor volume was reduced to 25 cc by use of properly designed cores. The vessel was equipped with a rupture disk, pressure gauge, and internal thermocouple. A 2000-watt double-wound heater surrounded the reactor. The inner winding of the heater was controlled automatically by means of a 'Guardsman" temperature controller. The outer winding was kept at a constant current input by manual adjustment. The outlet of the reactor was connected to two receivers installed in parallel to permit separate collection of the product obtained during steady state. Gases formed during cracking were released from the receiver by means of a back-pressure regulator. The gas was collected by displacement with water and measured, depending on its volume, either by a wet test meter or by the amount of water displaced. Transfer lines from the pump to the radiation cave and from there to the receiver permitted the installation of the reactor in the source. These transfer lines were 1/4"-OD, 1/16"-ID stainless-steel high-pressure tubing and were 51 feet long. They were wrapped on the outside with properly insulated No. 20 nichrome wire to avoid freezing of cetane in the lines. A flow diagram of the system is shown in Fig. 14, IF- z 0 0I I OURCE INSD _ DJ A DIN1:-GS ( BPC MET FD TOIN GASIDE RUSKA PUMP Fig. 14. Continuous flow reactor. A typical continuous run was performed in the following manner. The reactor was brought up to temperature and controlled at the temperature indicated by the internal thermocouple (T.C.-5 or T.C.-2), which was placed in the thermowell at the center of the reaction zone. The system was pressurized to 400 psig with N2 and, when all temperatures were lined out, the feed pump was|

i. The University of Michigan * Engineering Research Institute sta:ted. Pressure, temperatuire, feed-rate, and gas-meter readings were taken at regular fifteen-minute intervals. The liquid product was collected in the discard receiver. When process conditions had reached steady state, as evidenced by the constant rate of gas production, the liquid product was diverted to the second receiver for collection of the official period product. The period required to attain steady state in blank runs was of the order of two hours. A longer period, generally three to four hours, was required to obtain steady state in irradiated runs, due to the increased volume of the longer transfer lines. The steady-state period averaged six hours, except where pronounced coking made impossible the continuation of the runs. The same reactor, heater, and thermocouples were employed for all runs. Temperature pro#files of the reaction zone were taken during each run to establish the temperature distribution in the reactor. At the end of each run the liquid sample was collected in a 250-ml flask cooled in dry-ice-acetone mixture. The flask was then fitted with a reflux condenser, the outlet of which was connected to a gas bottle. The lowpressure gas was thus allowed to escape overnight from the solution standing at room temperature. The unit was completely drained and all liquids were weighed to obtain an overall material balance. Gas samples for the run gas, blow-down gas, and weathered gas obtained in depressurizing the unit were collee ted. 2. Results.-The results of 13 runs performed in the manner described above are presented in Table VII. Here the average temperature is given as an integrated average of the temperature profile. Values of temperature along the reaction zone are also tabulated. The contact time is defined as volume of reactor volume of liquid feed/hr where the volune of' the reactor was taken as 25 ml. This volume was calculated from the actual dimensions of the reaction zone, which was 3-1/8 inches long. When the reactor was filled with water, the total volume was found to be 32.5 ml. The difference in volume was due to the void volume of the fitting and to dead space between the cores, the thermowell, and the walls of the vessel. It can be thought of as a preheater volume since the temperature of the feed through it was sensibly lower than the reaction temperature. The radiation dose was computed from Fig. 4, allowing a 58.6-percent absorption due to the reactor and heater surrounding it, as shown in calculations presented in the Appendix. Lower dose values reported in the monthly summaries of April and May, 1956, are due to erroneous calculations........ 20

TABLE VII CONTINUOUS PYROLYTIC CRACKING OF CETANE:~' SUMMARY OF RESULTS Run No. 3 4 67 77 8 9Y 10y 11 12 13y 14 157 16 Exp. Book Page No. 158786 158788 158791 158795 158793 158700 158702 158704 158706 158708 158710 158711 158712 D~t~ 4/2 ~/3 4/6 ~/n 4/~4 ~/~6 4/~8 4/2~ 4/23 ~/25 4/28 ~/3o ~/~ O Avg. React. Temp., ~C 503 452 452 449 427 425 510 474 510 473 547 547 547 '~ Temp. Along ReaqtQr Zone, ~C 0 in. (inlet) (1) 495 451 455 445 428 421 505 474 -- 466 536 536 556 1/2 im. 500 455 455 449 430 425 509 475 -- 470 543 543 543 ~' 1-1/2 in. (center)(1) 505 450 450 451 425 425 512 475 -- 475 551 551 551:$' 2-1/2 in. 508 450 449 449 423 425 510 473 -- 474 549 549 549 ~' 3 in. (outlet) (1) 507 448 449 449 422 425 507 471 -- 474 544 544 544 Corr. Avg. React. Temp., ~C (2) 503 452 449 445 427 421 506 474 510 468 547 543 547 Avg. React. Press., psig 400 403 404 405 401 400 406 401 408 402 403 402 402 Approx. Rad. Dose, krep/hr (3).... 42 42 -- 42 42.... 31 -- 42 -- ~0 Feed Rate, cc/min O. 380 0.380 0.380 0.380 0.380 0.380 0.380 0.380 0.380 O. 380 0.380 0.380 0.380 ~ ~-~ Contact Time, Reciprocal Hours 1.098 1.098 1.098 1.098 1.098 1.098 1.098 1.098 1.098 1.098 1.098 1.098 1.098 Duration of Run, min 270 375 375 390 360 360 345 360 195 345 135 255 165 rT1 Total Feed During Steady State, cc 102.2 142.8 143.3 149.3 137.2 137.0 130.0 136.4 74.1 131.2 51.1 97.1 62.9 Total Feed During Steady State, gm 78.1 109.3 109.8 109.8 106.0 105.2 100.1 104.2 56.5 101.6 59.2 74.8 48.4 Total Feed During Steady State, mole 0.346 0.484 0.485 0.485 0.467 0.464 0.442 0.459 0.'249 0.448 0.173 0.331 0.214 Liquid Product, gm 53.7 102.7 98.4 97.6 100.9 lO1.4 78.2 92.9 40.4 92.5 15.0 14.0 12.2 Liquid Lighter than Cetane, wt ~(4) 94 -- 25 27 16 9.0 72 59 74 50 87 93 96 ~I~ Liquid Lighter than Cetane, gm 50.5 -- 24.6 26.4 16.1 9.1 56.3 54.8 29.9 46-3 13.05 13.0 ll.7 '~, Off Gas, S.C.F. 0.269 0.156 0.140 0.152 0.137 0.130 0.262 0.176 0.198 0.163 0.276 0.367 0.330 Off Gas, gm 8.8 5.2 4.6 5.0 4.5 4.3 8.7 5.8 6.6 5.4 9.1 12.1 10.9 Weather Gas, S.C.F. 0.165 0.014 0.003 0.003 0.003 0.003 0.122 0.047 O. 164 0.019 0.065 0.034 0.079 Weather Gas, gm 7.8 0.5 0.1 O.1 0.1 O.1 5.7 2.2 7.7 0.9 3.0 1.5 3.7 fl) Moles of Total Gas/Mole of Feed. 1.50 0 421 0.353 0.382 0.358 0.343 1.04 0.595 1.74 0.487 2.36 1.45 2.28 fl) Material Balance, wt ~ 90 99 94 93.6 99.5 100 92.6 96.6 98.4 97.3 69 38.2 54.9 Total Product Lighter than Cetane, gm 67.1 -- 29.3 31.5 20.7 13.5 70.7 62.8 44.2 52.6 27.1 26.6 26.8 'I Cetane Converted wt %, output basis 95.5 -- 28.4 30.7 19.6 12 8 76.4 62 4 80.8 53 2 92.8, 96 4 98.1 ~... Liquid Product, wt ~, output basis 72.0 -- 23.8 25.8 15.3 8.6 60.8 54.5 54.6 46.8 48.2 47.1 43-6 Gaseous Product, wt %, output basis 23.5 -- 4.6 4.9 4.3 4.2 15.6 7.9 26.2 6.4 44.6 49.3 54.5. (1) Interpolated values. (3) These values were reported incorrectly in the April report~. ~' (2) This correction is made on the basis of Table VIII. (4) Gas chromatography determination from Standard Oil Company (Indiana).

The University of Michigan * Engineering Research Institute The product inspections were carried out by the Standard Oil laboratories and they consisted of mass-spectrographic analysis of the gases and of fractional distillation and gas chromatography of the liquid products. The product distributions in the gas samples were nearly the same in all cases. A typical analysis of the gas is reported in Table VIII. The molecular weight obtained from them was used to calculate the weight of the gas produced. Frac - tional distillation of the liquid products was not very conclusive and did not give a satisfactory comparison of the liquid products of irradiated and nonirradiated runs. Gas chromatographic determinations gave the percent unconverted cetane in the liquid and are reported in Table VII. These values, along with the gas-analysis data, were used to calculate the material balance and the product distributions in the manner indicated in the Appendix. The unconverted cetane, liquid, and gas distributions are shown in Figs. 15, 16, and 17, respectively. TABLE VIII THERMAL CRACKING OF CETANE GAS ANALYSIS Typical Composition of Off Gases Collected During Run Components Volume ~ Hydrogen 6 Methane 35 Ethane 25 Ethylene 7 Propane 13 Propylene 8 Butane 3 Butylene 3 Total 100 Avg. MW = 27.7 Typical Composition of Gas Obtained by Weathering at 25~C Components Volume % Methane 3 Ethane 17 Ethylene 4 Propane 38 Propylene 18 Butane 11 Butylene 9 Total 100 Avgo MW = 39.2 22

The University of Michigan * Engineering Research Institute 100 90 o 80 L G L_ j 70.-. - o Radiated - o o Thermal (No Ro 6' X uradiation) 0 60 o0 50 0 m 40 L2 Z 30 w 0 20 o 410 450 500 550 TEMP. ~C Fig. 15. Cetane present in product vs temperature. 100 90! 2 80 0 7- 70 60 0 Z I-r 4,30. Radiated 5 Fig. 16. Liquid-product Thermal (No radiation ) 0 20 23

The University of Michigan ~ Engineering Research Institute 100 90 80 U 70 0 0 Radiated, 60 0 Thermal (No radiation) z 30 520 <- 0 (410 450 500 550 TEMPERATURE ~C Fig. 17. Gaseous-product distribution vs temperature. 3. Thermocouple Calibration. —On several occasions du-ring the course ments, especially since two entirely different sets of ambient conditions were cave for radiation runsp During the colder months the temperature in the source cave was appreciably lower than in the laboratory. Repeatedly the temperature recorder was checked by immersing the thermocouples in boiling water. Also, all leads were checked by feeding to the recorder an emf generated by a portable potentiometer. No anomaly was noted in these tests, at least within the accuracy of the recorder scale (0.1-0.2 percent of full scale). Later a considerable amount of time was spent to check more closely the whole temperature -measuring system, since it appeared that a temperature difference occurred when the thermocouple connections to the extension wires in the source cave were held at temperatures lower than the corresponding connections at the recorder which was installed in the laboratory. Such conditi prevailed throughout the period in which continuous thermal cracking runs were conducted. In order to determine the extent of the error incurred, chromel-alrurel c~ 3 0 ~ a

The University of Michigan * Engineering Research Institute thermocouples were immersed in molten fixed-point substances. The temperature of the thermocouple-extension-wire connections was varied by immersing the connections in water maintained at constant temperature below the ambient temperature. The observed positive error is plotted in Fig. 18 vs the temperature difference between recorder-extension-wire connection and thermocouple -extension-wire connection for lead, antimony, and potassium hydroxide. Table IX gives approximate average temperatures for the laboratory and source cave for those runs carried out under the influence of gamma radiations. It can be observed that this correction is never more than 50~C The corrected temperatures are also reported in Table VII. 30 02o r o0 Lead (325"C) > > / / + Antimony (630"C) 0 [ In 0 I o K2Cr207 (397oC) 0O 10 20 30 TEMPERATURE DIFFERENCE BETWEEN RECORDER- EXTENSION WIRE CONNECTION AND THERMOCOUPLE - EXTENSION WIRE CONNECTION, ~C Fig. 18. Error in temperature measurement. TABLE IX TERMAL-CRACKING TEMPERATURE CORRECTIONS Run No.........Y 6 9 —_ —o7 1.y 15 Recorded Avg. React. Temp., oC 452 449 525 510 473 547 Avg. Source Temp., 0C 20 15 16 12 14 16 Avg. Room Temp., 0C 26 24 25 20 24 23 Temp. Diff, Between Junctions 6 9 9 8 10 7 Negative Temp. Error 3 4 4 4 5 4 Corrected 'Avg. React~ Temp., ~C 449 445 421 506 468 543 l l~~~~~~~~~2

The University of Michigan ~ Engineering Research Institute D. CATALYTIC CRACKING OF CETANE 1. Equipment and Procedure.-The equipment used for the catalyticcracking study is shown diagrammatically in Fig. 19. The reactor consisted of the standard stainless-steel reactor described in Fig. 7. The top of the reactor was connected to the "Ruska" positive displacement pump and to a nitrogen cylinder so that downflow of cetane and purge gas is obtained. The outlet of the reactor was connected to a 500-cc round-bottom flask which served as a product receiver. The receiver was immersed in an ice bath. The uncondensed vapors escaping from the receiver were further cooled by a salt-ice condenser. The vapors were then measured and collected in a gas bottle. The reactor was heated in an analogous manner as for the thermal-cracking experiments without catalyst. VENT FOR NITROGEN REACTOR nc l I WET-TEST d —I t, METER GAS L=i THERMO- L BOTTLE D C' EN COUPLESP I NITROGEN CYLINDER.LS CONDENSER CETANE-PRODUCT COLLECTOR ICE BATH DISPLACEMENT PUMP (RUSKA PUMP) Fig. 19. Catalytic cracking unit flow diagram. The catalyst was charged to the reactor in such a way as to form a bed at approximately the centerline of the reactor in order to obtain the most uniform temperature profile in the reaction zone. Two typical arrangements used in charging the catalyst are shown in Figs. 20 and 21. Before closing the reactor a glass-wool plug was introduced at the top in order to facilitate the vaporization of cetane. The reactor was then heated to a temperature of from 20~ to 30~C above the desired reaction temperature, while nitrogen at a rate of from 5 to 6 cubic feet per hour was passed through the reactor in order to dry out the catalyst. A charge of 250 cc of cetane was placed in the "Ruska" pump and the transfer line from the pump was filled with cetane up to the reactor. The run was started by turning on the pump and opening the valve at the upper extremity 26

The University of Michigan * Engineering Research Institute FINE QUARTZ COURSE CATALYST;CUSQUARTZ XGLASS WOOL I1 ' "I ':-1 \ 11 5.0..0. 813- 10511. D.3 " 7.69" 13.0" Fig. 20. Catalyst arrangement in reactor with spacer. 0.812" 1. D. COURSE QUARTZ ~~~0.75 ~" 3.6 1 13.0" 0.513" 0 513" Fig. 21. Catalyst arrangement in reactor without spacer. of the reactor simultaneously. Feed rate, temperature, and gas make were recorded at five-minute intervals. The run was continued until enough product was collected, usually one or two hours, depending on the feed rate used. At the end of the run, the catalyst was stripped with 2000 cc of nitrogen slowly introduced into the reactor. The liquid product was weathered at 650C for three hours. 2. Results.-A summary of the runs so far undertaken is presented in Table X. The overall average temperature was taken as the integral average of the temperature profile across the catalyst bed. No significant pressure drop was observed in any of the runs reported. The radiation dose was calculated as shownin theAppendix. The space velocity is defined as weight of catalyst Some data are lacking for Runs lc through s8c because these runs were made in order to gain familiarity with the equipment and to set up a rigid run 27

TABLE X CATALYTIC CRACKING OF CETANE Run No. lC 2c 3cY 4c 5c 6cy 7c 807 9c lOc 1107 12c l3cY 1407 1507 16c I~y 18 Date 5/24 5/25 5/28 5/29 5/31 6/2 6/4 6/6 6/1i 6/12 6/12 7/8 7/9 7/12 7/13 7/16 7/1 Depth of Bed, in. 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3. Average Temp., 'C 495 495 500 534 542 544 542 541 475 482 488 525 522 529 526 527 52 53. Middle of Bed Temp., OC 500 500 505 539 547 549 547 543 475 495 492 532 527 534 531 531 52 Bottom of Bed Temp., 'C 499 547 538 483 471 487 515 506 Top of Bed Temp., OC 477 536 533 440 487 487 523 521 Pressure Drop, psig 0 0 0. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Approx. Rad. Dose, krep/hr -- -- 42 - -- 42 -- 42 - -- 38 - 42 42 42 - 2 - Catalyst Charged, gm 7.75 8.27 7.59 8.17 7.96 7.82 8.01 8.11 25.02 24.56 23.13 25.20 25.54 25.20 25.20 25.20 25.2 52 Feed Bate, cc/hr 120 120 120 120 120 120 120 120 120 120 120 120 122.4 122.2 122.1 122.6 121. io Total Bun Time, hr 1.0 1.0 0.834 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1. 00 Total Feed, cc 120 120 120 120 120 120 120 120 120 120 120 120 122.4 122.2 122.1 122.6 121. Total Feed., gm 92.5 92.5 77.0 92.5 92.5 92.5 92.5 92.5 92.5 92.5 92.5 92.6 94.4 94.1 93.9 94.0 93. Total Feed, mole 0.409 o.409 0.340 0.409 0.409 0.409 0.409 0.409 0.409 0.409 -o.4o9 0.409 0.416 0.415 0.414 0.415 0.41 Space Velocity, w/hr/w 11.92 11.20 12.20 11.31 11.61 11.81 11.51 11.40 3.70, 3.77 3.68 3.68 3.70 3.74 3.72 3.73 3.7 Liquid Product, gm -- 93,5 55.4 92.0 -- - - - - 79.2 81.3 72.0 50.7* 81.4 62.15 68.25 67.2 Off Gas, S.C.F. 0.0188 0.0249 0.0 0.0560 0.0938 0.00 0.0758 0.0802 o.146o 0.1345 0.042 0.399 0.2985 0.279 0.280 0.362 0.29 Weathered Gas, S.C.F. -- - - - - - - 0.0111 0.0300 0.0143 0.0229 0.028 0.0121 0.0074 0.0313 0.0285 0.14 S.C. F. Total Gas/Gm-Mole Beed O.Q46 0.0649 0.0 0.165 0.230 0.0 0.185 0.223 04l 0.364 0.0663 1.058 0.746 0.690 0.750 094 072 Gm-Mole Gas/Gm-Mole Feed 0.055 0.0729 0.00 0.197 0.274 0.00 0.221 0.266 0.516 0.435 0.0793 1.265 0.894 0.825 0.898 1.125 0.90 Mat. Lighter than Cetane,% 7 8 6 9 12 10 12 9 Spill

The University of Michigan * Engineering Research Institute procedure. Considerable difficulties were encountered in maintaining constanttemperature conditions throughout the run due to irregular vaporization of the feed and to the comparatively short life of the catalyst, which does not permit a waiting period to reach steady state before the products are collected. The first problem was corrected by introducing a glass-wool plug at the top of the reactor to obtain a better distribution of the feed; the latter may be controlled by establishing a given temperature profile in the catalyst before each run is started. Results of the product analysis are incomplete at the present time. Gas chromatographic results reported as percent of material lighter than cetane are included in Table X for Runs lc through 8c. IV. INTERPRETATION OF RESULTS AND CONCLUSIONS A. GAMMA-RADIATION EFFECTS AT ROOM TEMPERATURE The various experiments in glass vials were intended to serve as exploratory tests at room temperature. As shown from Table I through IV, the vials were exposed to doses from 5 to 30 megarep. The irradiated vials were shipped to the Standard Oil Company (Indiana) for inspection. An analysis of the data revealed that the maximum conversion detectable was of the order of 0.5 percent. In light of these results, the experimental work at room temperature was discontinued. Bo BATCH EXPERIMENTS AT ELEVATED TEMPERATURE Attention was turned to batch experiments at elevated temperature. Although they required more elaborate equipment than glass vials, it was hoped that they would serve to determine a possible temperature threshold above which appreciable radiation effects could be detected. Pentene-l and cetane were selected as the most suitable starting compounds for this study, since the former would tend to polymerize at moderate temperature while cetane would tend to decompose. After various trials, a constant-volume technique, previously described, was selected as most suited for this work. Experiments with pentene-l from 1500 to 3000C showed that some polymerization was taking place even without irradiation, as revealed by reactor pressures lower than the corresponding vapor pressure; however, the results were difficult to reproduce. The only run conducted under gamma radiation resulted in a lower final pressure as compared to those runs where pentene was only heated. The small number of experiments precludes, however, any further conclusion. 29

The University of Michigan ~ Engineering Research Institute Cetane was exposed to heat and radiation in a similar manner at temperatures from 400~ to 475~C. Cracking was evidenced by reactor pressures muc higher than the corresponding vapor pressure of cetane. The pressure-time curves observed during a run consisted of an initial straight-line portion and a terminal curvature (see Figs. 12 and 13). No consistent correlation of the slopes of the straight-line portion of the curves was possible because of the seemingly erratic change from experiment to experiment, as shown in Fig. 22, probably due to different rates of total heat input. The end-point pressure at which further increases in pressure depart from a straight line seems to be fairly consistent for all the experiments, as shown in the plot of Fig. 253. In addition, there seems to be no difference between the amount of gas formed in the irradiated and nonirradiated runs when the total pressure after one hour at reaction temperature is plotted vs the reciprocal temperature as shown in Fig. 24. An inspection of the nature of the products of the reaction shows, principally for those runs at and above 470'C, a greater degradation of cetane, as evidenced by the amount of liquids present in the products stored at 320C (since cetane is solid at this temperature). On the other hand, the color of the product gives no clue to possible differences of the cracking reaction. 100 0 0 X X o Blank X Radiation cr I0 w W XX a. o CD 0.00(130.00135.00140.00145.00150 RECIPROCAL TEMP, I/~K Fig. 22. Slope of linear pressure increase vs reciprocal temperature for cracking of cetane....... 30

The University of Michigan ~ Engineering Research Institute 10,000 a Blank < \ x Radiation I 000 ~RECPROCAL TEMP,.of new reactor x _ of new reactor x Radiation Fig 4 Total pressre after one hour at reaction temperature vs reciprocal temperature for cracking of cete. 1~-J ~3::OC u~ Possi~~~~0blan 51

The University of Michigan * Engineering Research Institute The failure to obtain more concrete information from pressure-time behavior of the experiments is due to the complexity of the constant-pressure system. The possible factors that produce the various discrepancies are: (1) the possible surface effects due to the degree of oxidation of the metal surface of the reactor; (2) the presence of the liquid and gaseous phase in the reactor, which makes it difficult to differentiate between pressure effects due to vaporization and those due to reaction; (3) changes of rate of heat transfer from the walls of the reactor to the body of the fluid, which were seemingly affected even by slight changes in room temperature; (4) complexity of the reaction, so that increases in pressures due to the reaction cannot be directly translated into degree of conversion; (5) undetectable leaks; and (6) varying heating and cooling cycles. No information from chemical analysis of the products could be obtained since their quantity was too small. The experiments under reflux conditions described earlier gave more definite evidence in favor of the inhibition of gas formation by gamma radiation. The simplicity and excellent reproducibility of these experiments make their results hard to confute and completes the picture partially gathered in the batch-cracking experiments. C. CONTINUOUS THERMAL CRACKING OF CETANE The favorable results obtained in the batch experiments justified further study of cetane cracking. The failure of improving the reproducibility of the batch runs was judged to be mainly attributable to the impossibility of preheating the feed rapidly and uniformly and suddenly quenching the reaction products. This shortcoming could be effectively eliminated only with the use of the available continuous reactor. All the continuous runs were conducted at the same space velocity; the only variable changed was temperature. The radiation dose was also kept constant, except that radiation runs were compared directly with blank runs, i.e., runs conducted outside the source. The results of the 13 experiments are somewhat ambiguous, mainly because a temperature error was discovered, as discussed earlier. This error, however small, unfortunately casts some doubts on the results, especially in the face of only minor differences caused by gamma radiation. Figures 15, 16, and 17 give the complete summary of the continuous work. The points shown are corrected for temperature, yet a consistently lower gas conversion is noted for the radiation runs. This is better illustrated in Fig. 25, where moles of gas per mole of feed are plotted vs temperature. The result is in agreement with the evidence obtained in the batch work. 32

The University of Michigan ~ Engineering Research Institute 3.0 0 2.0 LL I 0 1.0 OW 0.5 0.0 Figo Thermal No Thermal Gamma The product analysis was not very complete for the liquid products. o 1.0 CJ _j 0 A ratRef lux exp. 0.01 280 320 360 400 440 480 520 560 TEMP. ~C Fig. 25. Thermal crackingnot be approximately 1-hr contact time, moles of gas per mole of feed vs temperatreis antiC. the results can be interpreted better whent can the chemical analyses of the gasons, In two cases the Podbielniack fractionation of the liquid was not very satisfactory, especially when condensation products heavier than cetane are considered. 32 the same contact time cannot be easily generalized. D. CATALYTIC CRACKING OF CETANE ficient to be analyzed in detail. The preliminary results indicate that for moles of gas per mole of feed than the irradiated runs. It is anticipated that the results can be interpreted better when the chemical analyses of the gas, 55

The University of Michigan ~ Engineering Research Institute APPENDIX A. SAMPLE CALCULATIONS FOR DOSE RATE Unshielded Radiation Intensity = Io Centerline of reactor from centerline of source: 7 inches Height of reaction zone: 3.5 inches symmetrical to the centerplane of the source The intensity at various planes parallel to the centerplane of the source from Fig. 4 is plotted as a function of the distance from the midplane x on Fig. 26. 80 a. Uw 70 az 60 0 50 -4 -2 0 +2 +4 X, DISTANCE FROM MIDPLANE IN INCHES Fig. 26. Integration for radiation intensity. 34

The University of Michigan ~ Engineering Research Institute The intensity Io(x) is integrated graphically between the limits x1 and x2 which represent the extremities of the reaction zone. This gives the average intensity as 1= X1 I (x)dx =75 kilorep Ioavg- x2 - x1 1 hour B. SHIELDING CALCULATION The heater consists of a mixture of ceramic materials and aluminum. Since the densities and shielding properties of ceramics are similar to those of alumi um, the heater is considered to be all aluminum for shielding calculations. The simple exponential absorption law is assumed as I = Io e P where I = intensity after shielding, Io = intensity before shielding, p/p = mass absorption coefficient, x = thickness, and p = density. Shielding due to heater: Io = 75 kilorep/hr p/p = 0.055 cm2/gm p = 2.7 gm/cm3 x = 1.85 cm I = 75 e0055 x 2.7 x 1K85 57 kilorep/hr Shielding due to reactor: Io = 57 i/p = 0.053 cm2/gm p = 7.8 gm/cm3 x = 0.625 cm I = 57 e0053 x 7.8 x 0.625 = 44 kilorep/hr The dose determined by ferrous sulfate dosimetry5 in the reaction zone was 42 kilorep/hr 35

The University of Michigan ~ Engineering Research Institute C. SAMPLE CALCULATIONS FOR RUN NO. 12 Average reaction temperature was determined by the integral of the temperature profile, i.e., the average temperature throughout the reaction zone. Off gas = 0.198 S.C.F. (SoC.F. at 60~F and 1 atmosphere) gin moles gin Off-gas weight = 0.198 S.C.F. x 1.197 gm x 277 gm moes 6.6 gm S.C.F. gm moles Weathered gas = 0.164 S.C.F. Weathered-gas molecular weight = 39.2, from gas analysis gm moles gm Weathered-gas weight = 0.164 S.C.Fo x 1.197 gm moles gm S.C.F. 39.2gm moles = 7.7 gm Liquid-product weight = 4o74 gm 54,7 gi Total product weight = 54.7 gm Total feed weight = 56.5 gm Material balance = 54k7/56.5 x 100 = 98.4 percent Liquid lighter than cetane, wt percent = 74 from liquid analysis Liquid product lighter than cetane = 40.4 x.74 = 29.9 gm Total product lighter than cetane = 29.9 + 6.6 + 7.7 = 44,2 gm Product distribution output basis: Cetane unconverted = 54.7 - 44.2 /54.7 x 100 = 19o2 percent Liquid product = 29.9/54.7 x 100 = 54.6 percent Gaseous product =(6.6 + 7.7)/54.7 x 100 = 26.2 percent

1000 -...I - 10,000 01 '- 0 -LJ ~/0/ Uj C')~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 000 _ ZOO"~~~~~~~~~~~~~~~" 1 — /1~~~~~~~~~~~~~~~~~~~~~~~~~~~~000i -100 -75 -50 -25 0 25 50 75 100 0 400 600 800 I 000 TEMPERATURE ~F:o

The University of Michigan ~ Engineering Research Institute 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 TEMP OC Fig. 28. Specific gravity of cetane vs temperature, ~Ce A.P*I Pro ect 41. - -8 58..

The University of Michigan * Engineering Research Institute REFERENCES 1. Martin, J. J., Anderson, L. C., and associates, "Utilization of Gross Fission Products," Progress Report 4 (C00-124), Univ. of Mich., Ann Arbor, Eng. Res. Inst. Project M943, March, 1953. 2. Martin, J. J., Anderson, L. C., and associates, "Utilization of Gross Fission Products," Progress Report 5 (C00-196), Univo of Mich., Ann Arbor, Eng. Res. Inst. Project M943, September, 1953. 3. Martin, J. J., Anderson, L. C., and associates, "Utilization of Gross Fission Products,! Progress Report 6 (coo00-198), Univ. of Mich., Ann Arbor, Eng. Res. Inst. Project M943, April, 1954. 4. Lewis, J. G., and Martin, J. J., "Promotion of Some Chemical Reactions by Gamma Radiation," Univ. of Mich., Ann Arbor, Eng. Res. Inst. Project M943-4, January, 1954. 5. Weiss, J., "Chemical Dosimetry Using Ferrous and Ceric Sulfates," Nucleonics 10, No. 7, pp. 28-31, 1952. 6. National Bureau of Standards Bulletin 54,!'Protection Against Radiations from Cobalt-60 and Cesium-147,"' Washington, D. C., 1954. 39