A Study of Certain Fuels for Use in Jet-Propulsion Equipment Citation Hough, Eldred Wilson (1943) A Study of Certain Fuels for Use in Jet-Propulsion Equipment. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/cbz4-f916. https://resolver.caltech.edu/CaltechTHESIS:04232025-165230928 Abstract No abstract. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Physics) Degree Grantor: California Institute of Technology Division: Physics, Mathematics and Astronomy Major Option: Physics Thesis Availability: Public (worldwide access) Research Advisor(s): Unknown, Unknown Thesis Committee: Unknown, Unknown Defense Date: 1 January 1943 Record Number: CaltechTHESIS:04232025-165230928 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:04232025-165230928 DOI: 10.7907/cbz4-f916 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 17169 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 29 Apr 2025 22:23 Last Modified: 29 Apr 2025 22:24 Thesis Files PDF - Final Version See Usage Policy. 35MB PDF (Figure 16. Continuous Reaction Apparatus) - Supplemental Material See Usage Policy. 1MB Repository Staff Only: item control page

.A STUDY OF CERTAI E FUEL S FOR USE H J J ET-PR OPULS I OlJ EQ}J I PMENT Thesis by Eldre d W. Hough I n Part i a l Fulfil l men t of the Requir emen ts f or t he De gr ee of Do ct or of Ph il os ophy Calif or nia I ns t itut e of Te chno l ogy Pasadena , Calif ornia 1 943 A STUDY OF CERTAIN FUELS FOR USE IN JET-PROPULSION EQUIPlv!E:NT Thesis by Eldred W. Hough UNCLASSIFIED Classification changed to _____________________ __ _____________ Authortzy ~ _, _~ , ,----~ -4 ---_f:iq/_c;:,,-:_.,;__,_ Date~ o / 4 - C In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1943 TABLE OF CONTENTS Page List of Tables and Figures Introduction Part I. The Reaction of Nitric Acid and Gasoline . • . . . . . A. Bo The Reaction at Atmospheric Pressure, Utilizing Premixing . . . 2 The Reaction of the Preheated Fuels at Atmospheric Pressure o • ...... 5 A Study of the Reaction from Full-scale Jet-propulsion Equipment • • 14 ..... 6 c. Part II. 2 • 0 0 ♦ e ♦ e ..... . ...... ... . D. The Initiation of the Reaction . . . . . . . . . • E. Summary of the Nitric Acid and Gasoline React.ion . 25 . . 30 o 30 The Reaction of Nitric Acid and Aniline • • . . . 18 A. Crucible Tests with Nitric Acid and Aniline B. Rate of Reaction Experiment . .. .. . . .... 32 C. An Analysis of the Product s of Reaction of Nitric Acid and .Aniline. . . . . . . . . • . . . • . . • 35 ..... ................ 39 The Lowering of the Freezing Point of Aniline by Addition of Certain Compounds • . . • . . . . . . A-1 The Measurement of the Bubble-point Pre ssure, Specific Weight, and Viscosity of the Reactants • B-1 The Ignition of Aniline-Acetone-Air System. . • . C-1 D. Appendix A. Appendix B. Appendix C. Summary e • LIST OF FIGURES Apparatus* 1. Schematic Drawing of Apparatus for the Premixed Reaction of Nitric Acid and Gasoline. 2. Details of Mixing Chamber for Liquid-phase Reaction Equipment. z. Gas-phase Reaction Apparatus for Nitric Acid and Gasoline. 4. Equipment Used for Sampling the Products of Reaction. 5. Arrangements of Equipment for Gas-phase Reaction Investigation. 6. Photograph of Gas .Analysis Apparatus. 7. Flow Sheet of Apparatus for Ignition of Nitric Acid and Gasoline Using Opposed Injection Nozzles. 8. View of Equipment Used to Supply and Measure the Reactants. 9. Vi ew of Heat Transfer Equipment. 10. Arrangement of Heat Transfer Equipment. 11. Arrangement of Reaction Chamber and Pressure Control Equipment. 12. View of Ignition Chamber Using Opposed Jets. 13. Arrangements of Ignition Equipment in Reaction Chamber. 14. Sketch of Rate of Reaction Equipment. 15. Rate of Reaction Equipment. 16. Continuous Reaction Apparatus. 17. View of Continuous Reaction Chamber and Control Valve. A-1. Schematic Drawing of Apparatus Used in Determination of Freezing Point. B.-1. Bubble-point Pressure and Viscosity Equipmento B-2. Schematic Drawing of Bubble-point Pressure and Viscosity Apparatus for Corrosive Liquids. C-1. Diagram of Apparatus (Used for Study of Ignition of AnilineAcetone-Air System). * Figures containing description of apparatus ar e referr ed to by arabic numeral s. LIST OF TABLES I. Summary of Results for Minimum Preheating Temperature for the Gas Phase Reaction of Nitric Acid and Gasoline Reaction at Atmospheric Pressure. II. Composition of the Products of Reaction of Red Fuming Nitric Acid and Gasoline. III. The Reaction of Nitric Acid and Gasoline at Atmospheric Pressure. IV. Composition of the Products of Reaction of Nitric Acid and Gasoline with Full-scale Jet-propulsive Equipment. Vo The Spark I gnition of Nitric Acid and Gasoline with Opposed Jets. VI. The Hot Wire Ignition of Nitric Acid and Gasoline with Opposed Jets. VII. The Hot Wire Ignition of Nitric Acid and Gasoline with Jets Intersecting at 90°. VIII. Composition of Products of Reaction of Aniline and Red Fuming Nitric Acid. IX. Calculated Composition of the Products of Reaction of Aniline and Red Fuming Nitric Acid at Equilibrium. X. Composition of the Products of Reaction of Aniline, Acetone, and Red Fuming Nitric Acid. XI. Calculated Composition of the Products of Reaction of Aniline, Acetone, and Red Fuming Nitric Acid at Equilibrium. XII. Composition of the Products of Reaction of Orthotoluidine and Red Fuming Nitric Acid., XIII. Calculated Composition of the Products of Reaction of Orthotoluidine and Red Fuming Nitric Acid at Equilibrium. XIV. Composition of the Products of Reaction of Aniline, Orthotoluidine, and Red Fuming Nitric Acid. XV. Calculated Composition of the Products of Reaction of Orthotoluidine, Aniline, and Red J!uming Nitric Acid at Equilibrium. A-I. Freezing Points of Mixtures of .Aniline and a Number of Additive Agents. LIST OF TABLES (cont.) B- Ia. Volumetric Phase Behavior and Viscosity of Fuming Nitri c Acids,. B-Ib. Compositions of Fuming Nitric Acids. B-II. Bubble-point Pressure of .Aniline-Acetone Mixtures. B-III. Visc osity of .Aniline-Acetone Mixtures Containing 0.2975 Weight Fraction Acetone. C-I. Ignition Limits for the Gas Phase of Heterogeneous Mixtures of Aniline, Acetone, and Air. C-II. Calculated Composition of Gas Phase and Ignition LilI'i ts of Aniline-.Acetone-Air Mixtures. LIST OF FIGURES Data* I. Time Delay Associated with the Cricible Ignition of Nitric Acid and Aniline at 77° F. II. Ignition of Aniline-Acetone-Nitric Acid Mixtures. III. Ignition of Aniline-MP,thanol-Ni tric Acid Mixt1.1.reso IV. Adiabatic Reaction Temperature and Specific Impi;1.lse Obtained from Reaction of Aniline aJld Red Fuming lifi tric Aci,i, Asswning Thermodyn~~ic Equilibrium. V. Calculated Equilibrium Composition of the Reaction Products of Aniline and Red Fuming Nitric Acid. YI. Reaction Temperature of the Aniline-Acetone-Red Fuming Nitric Acid System. VII. Influence of Mixture Ratio upon Flame Temperature in the Orthot oluidine-Red Fuming Ni tric Acid System. VI II. Influence of Mixture Ratio upon :iname Temperature in the Orthotoluidine-Aniline-Red Fuming Nitric Acid System. A-I. Freezing Points of Anili ne and a Nwnber of Additive Agents. A-II. The Freezing Points of Aniline-Orthotoluidine System. A-III. The Freezing Points of Aniline with Various Compounds. J3-I. Bubble-point Pressure of Fuming Nitric Acids. B-II. Specific We ights of Puming Nitric Acids at the Bubble Pointo :B-III. Visc os ity of Fuming Nitric Acids at the Bubble-point Pressure. ]-IV. Viscosity of Aniline-Ac etone Mixture Containing 0.2975 We ight Fraction Acetone at t he Bubble Point. C-I. Experimental and Calculated Pressure Limits of Inflarnma.bili ty for .Aniline-Acetone Mixtures in Air at Various Temperatures. C-II. Critically Chosen Pressure Limits of Inflammability for AnilineAcetone Mixtures in Air at Various Temperatures. G-!II. The Index of Refraction of Aniline-Acetone Mixture at 78.5° F. * Figures containing description of data are referred to by Roman numerals. INTRODUCTION The work presented here is a part of that accomplished during the last two years in the California Institute of Technology Department of Chemical Engineering in connection with the development of jetpropulsion equipment by the Guggenheim Aeronautical Laboratory of the Institute for the use of the United States Army Air Corps. The· primary object has been to collect fundamental data regarding the behavior of fuels which would aid in the design and use of jet-propulsion equipment. Wherever possible, calculation of the behavior has been made for comparison with experiment with a view to eliminating the necessity for future similar experiment. The thesis consists of descriptions of methods and apparatus used, tabt1lation of results, calculated results when such calculations were possible, and conclusions regarding the behavior of the fuels in contact. This infonnation is used at present primarily in the desig-n of the jet-propulsive unit itself, i.e., injector, reaction chamber, and nozzle. The appendices contain information concerning the properties of individual fuels. These data are used in consideration of storage, ex- pulsion of fuels from storage, and the transfer of heat from the jet w1it to the incoming fuels. PAP.ii I. A. THE REACTION OF NITRIC ACID AND GASOLINE The Reaction at Atmospheric Pressure, Utilizing Premixing. Purpose_: In thl.s work it was desired to ascertain, by visual observation of the reaction between the two fuels, first , the effect of mixing the reactants, and , second, the effect of mixture ratio.• Apparatus: nsed. Figure 1 constitutes a schematic drawing of the apparatus The acid was stored in the glass flask ! and flowed to the apparatu.s throngh the resistance la• The pressure drop across the resistance was indicated by the difference in l evel of the acid in the arms of the manometer~. the top of which was connected to the top of the storage flank by means of the tube 12.• The gasoline was stored in a similar flask I and admitted to the apparatus through the resistance E.• Two mercury-in-glass manometers Q. and!! were provided to measure the pressure in the gas space above the fuels. The gas volumes above the two fuels were not connected in later measurements. Provision was made to add resistance to the res- pective fuel circuits by inserting glass rods il. and K into the glass tubing just below the storage tanks for the purpose ef varying the relative rates of flow of the two liquids. The absolut e rate of flow was varied by varying the pressure on the fuels in the storage vessels, using the regnlat or l!, connect ed between the fuel system and a nitrogen tank. Figure 2 illustrates the details of the premixing injector. Gasoline entered through the tube M, and was forced upward, along with the acid which entered in the tube 9., through the small annular space between the outside of a glass rod li and the inside of a glass tube f and into the -* Mixture - ---- ---- --- - - ---- --- --ratio was taken as the number of po1.1nds per second of acid used per pound per second of gasoline. - 3 nozzle B.• The two fluids were mechanically mixed by rotation of the rod!, the lower end of which was immersed in mercury and connected by means of a short piece of rubber tubing to a steel shaft!, which was connected to an external pulley through a packing gland. Since the clearance between! and!: was quite small, about 0.01 inch, and the rod f could be driven at speeds up to 500 revolutions per minute by a small electric motor, it is certain that the two nearly immiscible fluids were well mixed mechanically. Cooling of the mixing chamber was provided by the water jacket.§_. The nozzle 1i of Figure 2 was placed through a rubber closure into the bottom of a steel tube! of Figure 1 which was provided with three small observation ports and was exhausted to the atmosphere by means of a suction fan. Water .was circulated on the inside surfaces of the tube to remove unreacted fuels and to prevent excessive corrosion from the presence of free nitric acid. Frocedure: The mixture ratio was adjusted, the flow started, and the reaction then initiated by means of an oxygen-gas flame inserted through an observation port. The mixture ratio was adjustable during a run ae was the mixing, the latter by varying the motor speed. Results: Two different grades of nitric acid were tried: technical and red fuming. The first exhibited a specific weight of 83.c28 pounds per cubic foot at 77o F, which corresponds to 0.465 per cent water according to reference (1). The specific weight of the second was 98.65 pounds per cubic foot at 77° F, which corresponds to 0.19 weight fraction nitrogen dioxide if no water is present, according to Klemenc and Rupp (2). One grade of gasoli ne was used, namely, summer-grade unleaded automotive fuel (sample 210) purchased from the Texas Company. This fuel had a specific weight of 46.269 pounds per cubic foot at 80° F and a carbon-hydrogen ratio which can be expressed by the f onnula c8H16 • 18 . nitric acid behaved in much the same manner. Both samples of The mixture ratio was varied between the limits of 4 to 1 and 10 to 1, and the flow rates were about 1.8 x 10-3 pounds per second of total reactants, which, with the glass rod E of Figure 2 turning at 500 revolutions per minute, gave a well-dispersed emulsion in the tip. The dispersed mixture was ignited by means of a small oxygen-gas name mounted to come in contact with the fluid in the immediate vicinity of the tip of the capillary nozzle R shown in Figure 2. The reaction, at atmospheric pressure, did not occur at any time in the immediate vicinity of the tip. In the region above the tip the more intense burning was on the surface of the droplets, which burned more rapidly as their size decreased upon falling through the surrounding blue flame of the gasoline reacting with the air. This behavior indicates that the reaction is of the gas-phase type* at atmospheric pressure since, even with good liquid-liquid dispersion, rapid rates of reaction do not occur. In the case of full-scale equipment utilized by other parts of this project (3), the probable reason for the more violent reaction experienced, if the reaction is of the gas-phase type at higher pressures, is the increased rate of heat transfer to the droplets of liquid from the gas, which is known to be approximately proportional to * In this dissertation, 11 gas phase 11 , referring to the reaction between two liquids at a given pressure, will be used to indicate that a large amount of heat transfer into the reactants, presumably enough to cause their vaporization, is a much more important factor in obtaining energy from the reaction than is mixing. "Liquid phase" will be used to denote the opposite type of reaction; that is, the reaction between two liquids at a given pressure is of the liquidphase type if the intimacy of contact of the liquids, referring specifically to interfacial area per unit of t otal volume. is of more importance in obtaining energy from the reaction than is heat transfer into the liquids. the pressure (4) for constant-temperature difference between the gae and the dropletse B. The Reaction of the Preheated Fuels at Atmospheric Pressure. Purpose: The experiment descriood previously, which indicates that the reaction between nitric acid and gasoline does not occur to an appreciable extent in the liquid phase at atmospheric pressure, suggests the investigation of the reaction .with pr~eheating.. This was done, with the object of determining the minimum preheating temperatures of the fuels necessary to give reasonably rapid reaction in the cases of several grades of nitric acid, and of ascertaining the nature of the products of re~ction as a function of mixtu.r e ratio in the case of the red fuming type of acid if the preheating temperatures were held constant at these minimum values. Apparatus: The equipment used for the study of the gas-phase reaction is shown in Figures 3 and 4 and is shown in schematic representation in Figures. The flow of the two fluids from the storage flasks was controlled and measured in the same manner as for the liquid-phase apparatuse The nitric acid was introduced into a primary preheating section!. in Figure 5 of large cross-section compared to the line before it. Thus the acid vapors were permitted to escape without disturbing the flow of the liquid. The vaporized and partially decomposed acid was then introduced into a preheating section~. which raised the temperature of this material to the desired value. A thermocouple in the well! was provided to measure the temperature of the preheated oxidizing agent before it entered the reaction chamber~. The preheaters ! and !t were constr~cted of pyrex glass tubing, with resistance wire wound on the glass, and were operable up to 750° F. There was sufficient heat capacity to avoid appreciable variations in outlet temperature due to minor variations in flow rate. The gasoline was preheated in a similar fashion, but in a single preheating chamber .Q., and its temperature was observed by a thermocou.ple in the well!• The reaction was initiated by means of a spark plug located at ,li, and the products were removed through a trap II. by a suction fan attached to l• The pressure existing within the chamber could be ascertained by a water-in-glass manometer 1• Product samples for analysis •ere drawn off through a heated filter M., through a condenser li, at 60° F, and then through a trap Q, a condenser t, and a second trap g., all at 32° F. From the trap~. the gas flowed into the storage reservoir R, which had been evacuated previously through the line~. The pressures existing in the different parts of the system were ascertained by two mercury-in-glass manometers!• of gas were withdrawn for analysis through the stopcock u. Samples The water con- densing in the traps Q and~ was measured gravimetrically, and the acid present was titrated. These data together with the t otal quantity and ana- lysis of the gas collected in 11 permitted the evaluation of the water content of the products of reaction. The container ,R was filled at the rate of about 1 cubic foot per hour, maintaining the vapor pressure of water in the container essentially the same as that in the second trap, about 2 millimeters of mercury, which made the quantity of watering_ negligible c~npared to the quantity in the traps Q. and 2.• The Orsat gas-a..~alysis apparatus used in these investigations consisted of a manifold with various absorption pipettes connected through stopcocks and of a constant-volume burette connected directly to the manifold. A mercury-in- glass manometer measured pressure in the burette relative to the atmosphere. Constant temperature in the measuring burette was maintained by means of a - 7 - water jacket. The sample of gas to be analyzed was passed successively through solutions of sulfuric and nitric acids, potassium hydroxide, potassium pyrogallate, and cuprous chloride to determine, respectively, the relative quantities of oxides of nitrogen, carbon dioxide, oxygen, and carbon monoxide. In one case (0.159 weight fraction gasoline) the relative quantities of nitric oxide and nitrogen dioxide were measured by selective absorption in concentrated su.J.furi~ acid and in concentrated sulfuric acid with concentrated nitric acid. Frorn the determination of the total nitrogen and the trivalent nitrogen in the first solution, and of the trivalent nitrogen in the second solution, the oxides of nitrogen could be calculated. The total nitrogen was determined by nitrometer, while the trivalent nitrogen was determined by titration with potassium pennanganate. It was not possible to determine hydrogen or hydrocarbons with this apparatus. An improved Orsat apparatus was constructed for later measurements. An absolute-pressure manometer and a calibrated burette in a constanttemperature air bath are used for measurement of the quantity of gas. A copper oxide combustion tube and a platinum wire combustion pipette have been added for the determination of carbon monoxide and hydrogen, and of hydrocarbons, respectively. Figure 6. This apparatus is Ehovm in the photograph, The air bath containing the measuring burette is shown at the right of the apparatus, and the electronic circuits associated with the absolute-pressure manometer and the air bath are in the box near the center of the apparatus. bu.rette. The absolute-pressure manometer is above the measuring - 8 - Procedure: The vaporizing and preheating tubes!,!, and .Q. were brought to the desired temperatures. Then a small stream of air was introduced into the reaction chanber through a stopcock (not shown) between the manometer!!, and the reaction chamber Q.. The spark Kwas started. iToon gasoline wa! admitted to the chamber Q:. through the tube .Q., a gasoline-air reaction Acid was admitted through! and l, and the air was stopped as resulted. soon as the gasoline-acid reaction started. The suction fan was adjusted so that the reaction chamber was at a pressure of a few inches of water above the atmospheric pressure. Then a sample of gas was drawn through the condenser and trap sequence li ~ f ~ into the evacuated glass battle li• The time involved in starting the reaction was about 10 minutes, after which time the lower part of the steel reaction vessel maintained a steady temperature of about 1300° F. 20 minutes. The withdrawal of the sample required another During all this time the reaction proceeded smoothly, and no difficulty was experienced with respect to either instability or explosion. Results: Three different samples of nitric acid of different specific weights were tried with the same sample of gasoline previously described in order to establish optimwn preheating temperatures. The first sample of nitric acid, No. 213, was of the c.p. red fwning variety, and exhibited a specific weight of 97.965 pounds per cubic foot at aoo F, corresponding to 0.170 weight fraction nitrogen dioxide. This reacted with the gasoline described in the liquid-liquid experiment (sample 210) to give the hottest flame, using the minimwn preheating temperature of 490° F for the acid and 490° F for the gasoline, at weight fraction of 0.875 red fwninve acid in the fuel. As a matter of interest simple equations were com- puted for this particular sample of acid, the composition of which is derivable - 9 from the density if no water is present (2), and for this gasoline, assuming a molecule that would give the correct carbon-hydrogen ratio. If all the products are assumed to be carbon dioxide, water, and nitrogen, the simple reaction indicates 0.8417 weight fraction red fuming nitric acid in the reactants. Taking the products to be carbon dioxide, water, and nitric oxide, the equation corresponds to a weight fraction of red fuming nitric acid of 0.9014 in the fuel. It is realized that the products assumed here are quite different from the true ones, but this type of calculation seems to have definite value in estimating the mixture ratio to be tried first when experimenting with or making more exact calculation for an unknown fuel. In this case the experimental value of weight fraction nitric acid which gives the best reaction with the lowest preheating temperature lies between the two calculated values. The second sample of nitric acid tried, No. 214, was of the commercial fuming variety and was fowid to have a specific weight of 92.081 powl.ds per cubic foot at 77° F, which corresponds to 0.156 weight fraction water in the acid. With sample 210 of gasoline the optimum pr eheating temperatures were found to be 530° F for the acid and 460° F for the gasoline, the corresponding weight fraction of white fuming acid in the fuels being o.815 0 Ha.vever, a somewhat more intense reaction was obtained at ~ weight I fraction of 0.844 fuming acid in the fuels, the acid being preheated to 7000 F. This reaction was not so intense as that of the red fuming acid. Simplified equations give 0.9650 weight fraction fuming nitric acid in the reactants if the products are assumed to be carbon :dioxide, water, and nitrogen, and 0.9129 weight fraction acid in the case of carbon dioxide, water, md nitric oxide. - 10 - The third sample of nitric acid, No. 209, possessed a specific weight of 83.528 pounds per cubic foot at 80° F, which corresponded to 0.465 weight fraction water in the acid. It was not possible to sustain a reaction of this material and sample 210 of gasoline, even though both reactants were preheated to 750° F. This behavior bears out that predicted by the other two samples of acid. The very deleterious effect of water is probably due to the large amount of energy required for its dissociation (5). The results of these essentially qualitative experiments are presented in Table I. The products of reaction of the sample of commercial red fuming nitric acid, No. 217, of specific weight 98.459 pounds per cubic foot at 80° F, which corresponds to 0.1905 weight fraction nitrogen dioxide if no water is present, and of the sample 210 of gasoline used earlier were determined as a function of mixture ratio at the optimum preheating temperature of 480° F for both reactants. The actual rate of flow, in the case of 0.159 weight fraction gasoline, was 3.4-8 x 10-4 pounds per second and 0.66 x 10-4 pounds per second for the nitric acid and gasoline, respectively. The un- certainties in these rates are about 0.033 x 10-4 pounds per second. The composition of the products for three different weight fractions of gasoline in the mixture is indicated in Table II. The amount of carbon monoxide in the products varies markedly with mixture ratio, being less than 0.01 mole per cent at 0.120 and 0.159 weight fractions of gasoline but increasing to 17.30 mole per cent at 04266 weight fraction gasoline. Also the quantity of oxides of nitrogen present at this pressure is quite small in the case of 0.266 weight fraction gasoline, but is larger for the mixtures more rich in acid. Carbon dioxide is higher in the case of the lower weight fractions of gasoline. No free oxygen was found in any case. - 11 - It is possible to estimate the amount of free carbon produced in the reaction from material balance. As a matter of interest this has been done, and the results expressed in Table III. ]y material balance, the relative amounts of hydrogen and nitrogen in the gas inert tottle early Orsat analysis apparatus described above were estimated. It is seen that the carbon and hydrogen increase if the weight fraction of gasoline in the reactant mixture is increased. The enthalpy change of reaction assuming reactants and products at 64° F has been calculated from the data of the ]ichowsky and Rossini (5) relating to the molal enthalpies of the components, and is shown in Table III. It is seen that the energy obtained, per pound of reactants, does not vary greatly as the mixture ratio is varied. This is to be expected since the enthalpy change for formation from elements in the molecular state of nitric acid, water, and carbon dioxide is high compared to carbon monoxide, hydrogen, and gasoline. Increasing the amount of acid, which is the oxidizer, gives more carbon dioxide and water which have high enthalpies of fonnation, but more energy is also required to decompose the additional nitric acid, res.tlting in a very small overall effect of mixture ratio upon energy released. The adiabatic temperature of the reaction has been calculated using the relation which assumes a thennally perfecta but calorically imperfectb gas: ---- --- --------- ---- - --- a A thermally perfect gas obeys the equation of state pv = bT 9 where pis the total pressure, vis the specific volume, T the absolute temperature of the gas, and b a constant. b A calorically perfect gas is defined as a thermally perfect gas with , constant specific heat under constant pressure or constant volume. - 12 - - .aH where C dt ~ tl H = enthalpy change per pound of reactants for both reactants and products at a temperature t 0 °F tc = chamber or reaction temperature in °F to = temperature of reactants in °F mh = weight fraction of component h present in products CPh = heat capacity of product h in Btu/lb/OF The values of enthalpy have been calculated, from spectroscopic transition probability a..~d energy level values of heat capacity available in the literature, by the Department of Chemical Engineering (6). These temperatures are recorded in Table III. The expressions for the equilibrium constants for several reactions between the products nave been computed on the basis of analytical results and compared with the equilibrium constants given by Lewis and von Elbe (7) evaluated at the adiabatic reaction temperature. In general, considerably more oxides of nitrogen and carbon are present than equilibrium values indicate. This may mean that equilibrium is not attained in the reaction, or that the equil ibrium shifts during the sampling process 9 or both. The nitrogen dioxide found was considered to be all decomposed to nitric oxide and oxygen at the reaction temperature for the purposes of calculation of experimental expressions. The symbol (h) will be used to denote the fugacity in atmospheres of the component ho The f ugaeities 01· the gases present were computed on perfect gas assumptions. - 13 - In the ease of 0.266 weight fraction gasoline of Table II. the experimental value oi the expression for the water-gas reaction, (OO)(H2o)/(co 2 )(¾)• is s.78, compared to 5.75 for the equilibrium constant at 3780° F. This is good agreement. The computed value of (00) 2/(00z)(C) is probably of the order of 10, while the equilibrium constant ha!! a value of 191,000. This is possibly satisfactory, because no actual determination was made of the carbon. The oxygen, computed from the equilibrium value of l the (CO)(Oz) 2/(co 2 ) expression,0.0166, comes out to be 0.000132 mole fraction, which checks because no oxygen was detected. The oxygen can also be computed ' from the equilibrium constant (H2 )(o 2 )~(H2o) which has a value of 6.0030. The oxygen computed on this basis is 0.00033 mole fraction. However, the check with respect to the nitric oxide, which can be computed from the equi1 l librium constant NO/(o2 )2(N2 ) 2 whose value is 0.046, and from the average of the two above indirect values for oxygen, 0.00023 mole fraction, is not satisfactory. The calculated value of nitric oxide comes out to be 0.000377 mole fraction, while the measu.red value ig 0.0545 mole fraction. The watei-gas reaction does not enter into the check for the two mixtures with more oxidizer, since no hydrogen or carbon monoxide was found. In the instance of 0.159 weight fraction gasoline, some oxygen i! present if the nitrogen dioxide is considered decomposed. l Using this amount ' to calculate the value of (N0)/(02) 2 (N2 )'i gives 3.30, which is much larger than the 0.066 indicated for equilibrium. oxide present. Again there is too much nitric The experimental and equilibrium values of (C0) 2 /(co 2 )(0) are, respectively, indetenninat e, because carbon monoxi~e was not detected and carbon was not mea,sured, and 300,000, which is possibly satisfactory. By l the expression (OOa)/(02 )2(00), which has a value at equilibrium of 0.050, and the above-mentioned value for oxygen, it is poeeible to calculate the - 14 - ..,alue of carbon .1monoxide if equilibrium ie obtained. This comes out to be 0.0806 mole fracti on and actually none is observed. Similarly it is possible to calculate the value for the hydrogen present from the equilibrium l constant (Hz0)/(o2 ) 2 (H2 ), which has a value of o.0087. The calculated value for hydrogen is 0.00237 mole fraction , against none observable by attempting to burn the gas inert to the Orsat analysis. This is considered to be only qualitative evidence that this equilibrium is not satisfied. For the 0.120 weight fraction gasoline point, there is again too much nitric oxide found experimentally. 1 l The value of (NO)/(o2 )2(N ) 2 2 computed from the analysis, evaluating the oxygen from the decomposition of the nitrogen dioxide as before, is 1.51. 1 The equilibrium value is o.oso. , The values of (CO)(Oz) 2/co 2 , (H2 )(022 )/(H2o), and (Co) 2/(co 2 )(C) are calculated to be, respectively, O, 0, and indetenninate, and the equilibrium values are 0 . 039, 0.0068, E.nd 300,000, respectively. ' These are considered to be in satisfactory agreement. C. A Study of the Reaction from Full-scale Jet-propulsion Equipment. Purpose: Since equilibrium does not obtain in the reaction at atmospherie pressure, it was desira~le to investigate the reaction at higher pressure, considering the possibility of calculating perfo nnance if equilibrium were actually established. The values of temperatures obtained from a gas analysis are of practieal importance in de·s ign consideration, and in addition it "IIOuld be possible to checl~ the efficiency of the wiit. Apparatus: The jet unit used was similar to that described in a report ( 3) of another section of this project, and in fact was designed, built, and operated by that section. - 15 - The chamber was 3 inches in diameter by 1 foot in length, inside dimensions, The jet diameter was 0.75 inch at the throat and 1.5 inch at the lip. A short ssnple tube with valve attached was inserted in a drilled hole leading into the jet which made an angle of 60 degrees to the direction of flow. The location of the inside end of the hale in the copper jet wall was halfway between the throat and lip. This tube connected to a container approximately 0.67 cubic foot in capaeity through two ice-cooled traps in series, all of which could be evacuated. The injector consisted of 36 holes 0.031 inch in diameter through which red fuming nitric acid entered, and of 18 holes 0.025 inch in diameter through which gasoline was injected. The fuel injector holes were arraneed on a circle about 2 inches in diameter, while half of the acid holes were on a circle about 0.25 inch less in diameter and half on a circle about 0.25 inch more in diameter, so that each stream of gasoline was impinged from two sides by streams of acid. Procedure: The sampling system was evacuated up to the valve near the jet. To insure the sample's being representative of the steady-state reaction the sample was taken during a 16-second time interval which omitted 3.5 seconds at the beginning and 3.5 seconds at the end of the run. The nitric acid used was of the commercial red fuming variety similar to sample 217 described earlier, and the gasoline was of the aviation type, 100 octane, leaded. Results: The first pa.rt of Table IV contains the results of the gas analysis. It is obvious that some air entered the sampling apparatus, because the presence of hydrogen is not compatible with that of oxygen. equilibrium is well satisfied here. The water-gas The value of (CO)(HzO)/(COz)(H2) from the gas analysis is 4.29, while that of equilibrium at this temperature is 5.9. - 16 - This indicates that the results are probably fairly reliable from the standpoint of calculation of performance , because a large part of the enthalpy change of the reaction is detennined by this equ.ilibriunt@ There is definitely nitric oxide in excess, since using the value f or the oxygen due te nitrogen dioxide and nitrogen tetroxide considered decomposed, pl us that found in the free state at room temperature, a value probably larger than actual, gives a vaue for (NO) / (o 2 )½(N2 )i of 1.592 compared to 0.051 for equilibrium. Certainly too much nitric oxide is present. The experi- montal values of (o2 )½(CO)/(C0 2 ) and (~)(o2 )½/(H2o) are both of abcut an order of magnitude larger than the equilibrium values, perhaps because of the presence of air in the sample. Unfortunately the experimentally measured values of thrust, flow of reactants, and chamber pressure are not available because of failure of the recording apparatus. However, the values calculat ed from the gas analysis agree fairly well with other similar runs. The main sources of error in the gas analysis are (i) the reaction of the oxides of nitrogen with the mercury used as a confining fluid in the Orsat apparatus , (ii) the solubility of the carbon dioxide in the sulfuric acid used as an absorbent f or the oxides of nitrogen, which was corrected for but may cause the detennination of carbon dioxide to be in error by as much as o.05 mole fraction according to the data of Markbam and Kobe (8), and (iii) the presence of hydrocarbons in the gas inert to the solutions used and to copper oxide at 482° F, which is probably small because this gas would. net burn in air. It is possible to test the consistency of the gas analysis by a relation between exhaust temperature, the angle between the direct ion of - 17 - flow and the direction of observable discontinuities in the jet discharge, the exhauet velocity, the average molecular weight, and the ratio of isobaric to isochoric heat capacity. The relation whose development is indicated elsewhere (9) is T where _ e - u2 sin2 o( M oRg Te= exhaust temperature on U = exhaust velocity, feet per second ex: = angle between direction of flow and observable disturbances in the stream M = average molecular weight ~ =ratio of isobaric to isochoric heat capacity of the gas R = gas constant in foot pounds per °F per pound mole g = acceleration due to gravity in feet per second per second Using the values obtained from the gas analysis U = 6014, M = 26.04 9 o = 1.20, ahd a value of d. obtained from a photograph of a similar run, ct= 240, the exhaust temperature is found to be 2145° F, with an uncertainty of about 300° F, due mostly to the uncertainty of 1.5 degrees in measuring d-.. This is in agreement with the value of 2175° F of Table IV. It is to be regretted that these measurements could not be repeated and made more eatisfactory from the standpoint of correlating the measured large-scale data to the gas analysis, and also from the standpoint of eliminating air from the sample. At this time the problem of ignition and of stability of operation was more pressing than the analysis of the efficiency of the operation of the jet unit. - 18 - Do The Initiation of the Reaction. Purpose: The previous investigation indicated that the reaction between nitric acid and gasoline at atmospheric pressure is of the gas-phase type. This implies then that at higher pressures the reactants should be introduced separately into the cha~ber in the gas phase to eliminate irregularities that might occur in their vaporization if the energy for this process were supplied by the conduction and radiation from the hot gases of the reaction. However, objections to this procedure were raised by other sections of this project on the grounds that application of such a jet unit would be too complicated for practical use • .t\.n appreciable part of the difficulty connected with a u.,.~it that does not use preheating arises in starting the reaction. This process necessarily takes place near atmospheric pressure which is the pressure of poorest heat transfer from the gas to the liquide Hence it was decided to study the initiation of the reaction of the dispersed reactants with various mixt~re ratios and different degrees of dispersion, using different ignition devices. Apparatus: The schematic diagram of the flow of the reactants is shown in Figure 7$ The acid was stored in a welded stainless-steel tank fi and was maintained at a pressure of 20 pounds per square inch on the inlet side of a calibrated chemical proportioning pump[, by means of air pressure. The pump is of stainless-steel construction with Oa375-inch bore, a.stroke continuously adjustable from zero to 1.5 inches, and a speed of 59 cycles per minute. The discharge from the pump K entered a pressure vessel~. supported on a balance .Q.• The null point was indicated by electric contacts at 12,, which actuated small neon lamps through electronic circuits at E, and!, and - 19 .. was accurate to about 0a00l pound (0.5 gram). Nitrogen was introduced into the upper part of the vessel~ from a supply tank Q, and the pressure on the system was indicated by a gauge lo This chamber, in addition to indicating steady-state conditions of flow through the system, served to smooth the pulsating flow from the proportioning pump. An analogous syste~, designated by primed letters, was provided for the gasoline. The two pumps were linked mechanically. Figure 8 shows this apparatus with the proportioning pump in the right foreground, the controls in the left center, the balances with pressure vessels suspended below in the right center, and storage tanlcs at the far right. After leaving the vessel~ of Figure 7 9 the oxidant was passed through a heat exchange coil !I..• The coil, whose surface area in contact with agitated oil was 1.2 square feet, was quite large for the flow whim was less than 0.01 pound per second~ provided for the gasoline. The components of the heat e:x:chAAger are shown photographically in Figure 9. Figure 10. A similar heat exchanger !I,.1 was A cross-sectional diagram constitutes The fluid entered the device through the tubing A at the fitting! and passed downward through a tubing in contact with heated oil which was circulated by the impeller!• of the fluid. A valve l controlled the discharge The oil bath was heated electrically by chromel ribbon heaters wound on the outer wall of steel tubing. water-cooled packing gland Q.. provided for the gasoline. The impeller was driven through a A similar heat exchanger Q: 1 of Figure 7 was Platinum gaskets were used to seal the tubing to fitting joints where nitric acid was involved, while copper gaskets were used in the case of the gasoline. Provision was made to control temperatuxe - 20 - automatically by a thennoregulator l• As shown in Figure 7, the outlet of the heat exchangers connected directly to the reaction chamber through electrically heated lines. The products of reaction were rsnoved by an exhaust fan after being washed with a water spray in the scrubber unit.§.. The arrangement of one reaction chamber is shown in Figure 11, and a view of the component parts is shown in Figure 12. Referring to Figure 11, the reactants were introduced through small opposed nozzles ~ and Q. into the chamber A• The diame t ers of the different nozzles used varied from 0.0041 inch to 0.0168 inch. A frit t ed-glass filter,of opening approximately 0.0015 inch was used before each nozzle less than 0.010 inch in diameter. A pyrex window which also acted as a blow-out plug was provided in the fitting 11• A pressure release valve g which could be loaded by t he spring Kand adjusted by a hand wheel~ permitt ed the products of reaction to escape through the chamber M. to the exhaust port l• A stream of water was maintained in M. to remove any unreacted material, and a thennocouple was placed in I to detect ignition in the chamber. The chamber, which was of copper, was water cooled near the valve seat at the annular space li, and the valve ,ll was also cooled by wa.ter circulating in the interior. devices were introduced through the fit t ing!• cbangeable. Ignition Fi ttings.J: and~ were inter- A slightly different chambe r was construct,ea from stainless steel to avoid ionic corrosion from unreacted oxidant. The nozzles which injected the reactants were located at an angle of 90 degrees to each other, and a separate pyrex window rod blo~-cut plug were provided. made to heat the chamber electrically. Provision was Ignition devices were introduced into the chamber through a fitting at the top. This fitting was later - 21 - altered to accommodate a sample line for the study of the products of reaction of nitric acid and aniline. The chamber with this revised fitting will be shown later (Figures 16 and 17). The ignition devices used are shown in Figure 13. The fittings ~and~ utilize an electric spark, while the active element in.£. is a hot wire. The electric insulation of the fit ting~ was maintained in the presence of dispersed nitric acid by a continuous stream of oil introduced at Q. into the space~. The approximate energy of the spark was measured by a wattmeter in the primary of the spark transformer, and the resistance of the unit was ascertained by a milliammeter placed in series with the secondary. An improved fonn of this spark fitting is shown in Figure 13b. The effective dynamic oil surface was vertical, on the surface of the porous plug!, so that a smaller flow of oil was required to maintain the insulation of the plug. In the case of the hotc;Wire fitting .Q. no special insulation w~s necessary because low potential was used. The current in and voltage drop across the hot-wire element were measured at the temperature used in ignition attempts and at the ice point with a smaller current. This made possible a computation of an average temperature of the element and of the power dissipated per unit surface. All temperature measurements were made by copper-constantan thennocouples. The potentials were measured on a potentiometer. The thermocouples were calibrated against a strain-free platinum resistance thennometer which had been calibrated at the National Bureau of Standards. Final temperature measurements are probably within 0.5° F of the international platinum scale. - 22 - Pressure measurements were with Bourdon-tube gauges and were probably uncertain by less than 5 per cent. Tbis large value is acceptable for the present experiment because the ignition is experimentally found to be fairly insensitive to pressures in various parts of the system. The rate of flow of the reactants was ascertained from direct calibration of the pumps at various pressures and stroke lengths. The un- certainty was probably about 3 per cent, except for a short period when, because of corrosion of the pump valves, the uncertainty was much larger, possibly 10 per cent. Procedure: The heat exchangers, lines, and chamber were brought to the desired temperature. Then pressure was built up in the chambers~ and~•• In general a few seconds before the flow was started the ignition device was turned on by opening the valves between the heat exchangers and the reaction chamber. For short periods of flow it was found more accurate to calibrate the discbarge characteristics of the nozzles in order to ascertain the rate of flow of the reactants. The inside of the chamber was observed continuously through the observation port with a telescope and mirror arrangement to ascertain if reaction occurred or, in case no reaction occurred, if the injection nozzles were aligned. If the reaction did not occur, the flow of reactants was terminated after 10 seconds, the equipment flushed with nitrogen, and the process repeated. After each series of trials the whole apparatus was dismantled and cleaned to avoid ionic corrosion. Results: The influence of temperature, mixture ratio, and. dispersion was investigated along with the power input and, in the case of the hot wire, the surface density of power. - 23 - The gasoline used was of the unleaded aviation type, 89.5 octane, which had a specific weight of 44.261 pounds per cubic foot at 770 F, and was refined by the Shell Oil Company. The color was water white, and there was no noticeable gum formation from the fuel. The acid used was of the commercial red fuming variety, with specific weight between 97.350 and 98.€24 pounds per cubic foot at 770 F. If it is assumed that no water is present, tt~se values correspond to 0.137 and 0.190 weight fraction nitrogen dioxide, respectively. The results may be divided into two parts, (i) opposed injection nozzles using the copper chamber and (ii) injection nozzles intersecting at 90 degrees using the stainless-steel chamber. In the case of the opposed jets, spark ignition of about 15 watts power was tried first, at atmospheric pressure, with injectors in the range from 0.0063 to 0.014 inch in diameter, and injection pressure 75 to 100 pounds per sq11are inch above chamber pressure. with no air present gave no ignitions. Ten attempts at ignition Seventeen trials with air introduced at the start gave seven successful ignitions, which reacted smoothly after starting. These results are summarized in Table V. It is believed that the resistance of dispersed nitric acid and gasoline is so low that effective development of energy between the electrodes of the spark to ignite the mixture is prevented since, when the reactants were introduced 9 tr~ spark would gradually lose its intensity. However. it was by no means due to a failure of the insulation of the ignition fitting itself, because the discharge would recover to its original intensity when the flow of the reactants was stopped. This could be repeated a large number of times. - 24 - It was possible to st art the reaction with fair certainty by introducing oxygen at the beginningo The firs t reaction was gasoline~oxygen, initiated by the spark, which was of sufficient intensity to start the gasoline-acid flame when the acid was introduced. Opposed injection jets with hot-wire ignition were investigated. In general for average wire temperature greater than 1400° F, corresponding to a surface dissipation of energy of 390 watts per square inch, the ignition was satisfactory for mixture ratios of acid-gasoline lying between 4.5 and 9.2 if the wire was located so as to come in contact with a portion of the reactanh which are roughly in this range of mixture ratio. In this apparatus, the lower limit of hot-wire surface was about 0.05 square inch, smaller surface areas giving pressure surges because the probability of vaporizing enough reactants to start the reaction in a time before there is a large accwnulation of reactants in the chamber decrease,s with the amount of hot-wire surface. These results are indicated in Table VI. In general, oxygen or air was not necessary to start the reaction in this case. Nor was preheating essential. These facts indicate that the reaction,under the conditions of ignition in jet-propulsion equiµnent, is in the gas phase, since the hot wire, which gives successful ignition, tends to vaporize a large quantity of the reactants, while the spark treats more violently a smaller quantity. The ignition characteristics of the chamber in the jets intersecting at 90 degrees were ascertained. Careful alignment with water at the pressure of the reactants, as well as subsequent visual observation, insured that the intersection of the jets of reactants involved at least 75 per cent of the - 25 - cross-sectiona l area of each jet. A simple model, assuming that the two reactants mix and have i d.entic a l velocities a fterward, indicates tlia t 64 per cent of t he orig inal kinetic energy of the jets is used in t he mixing process in the c ase of opp osed jets but t ha t t his figure is reduc ed to 28 per cent in c a se of t he 90-degree intersection, if the mixture ratio in pounds of acid to p ounds of gasoline is taken a s 4 to 1. The time re quired for the dispe rsed re actants to traverse a g iven distance is only 30 per :cent less in the case of jets intersecting a t 90 degr ees. One would expect then tha t t he i gnition char acteri s tics wit h t he apparatus under consider at i on mi ght be markedly differ en t, because of p oorer dispersion. This difference was verified experimentally. In no c a se was i g nition sa tisfactory with t:be hot wire, which U.e previous e xpe ri me nts indicate is better than the s par k. ceeded. In one c ase , however, air was i ntroduced a nd the r eaction pro- I n one other ca.s e when the re a ct ants and cbamber were prehe a ted to 200° F, t he reaction proce e ded but was unst able. A l a r ge variety of ge ometric configur a tions of tbe hot wire were tried, and during tbe whole time the reactant jets were of velocity of t l~ order of 100 feet J:Br second. Table VII summarizes these r esults. E. Summary of t he Nitric Acid and Gasoline Rea.ction. All of the evidence collec ted indica tes t hat the re a ction of ni tric acid and gasoline as utilized i n t he various pieces of app aratus described is of the gas -phase -cype. On this hypot hesis i gni t ion is di ffi cult because, starting at a tmospreric :pr essure, heat transfer to the droplets of the reactants is probab ly at leas t one order of magnitude small er t han during t he operatbn - 26 - of the unit, if heat tr ansfer is considered proportiona l to t re pressure , for a g iven temper a ture. In large-scale practice when prehes ting is not employed, t he re a ction h a s been found to be i rJ1erently unst able. '.!.'his indic ct tes a time dela y between the introduction of the fuels into the chamber a.nrl their rea ction. Mec hanisms for two types of oscilla tions can be seen. is t he simple mechanic a l os cill a tion. The first A small rise in chamber pressure will r esult i n a drop of pressure across the injection orifice s, which will slow down t he r a te of injection of the reacta nts and c ause t he chamber pressure to drop. The lower chamber pressure will r esult in a l ar ger pre s sure drop a cross the injection orifices, which will g ive a sur ge of re a ct ants resulting in a hi gher chamber pressure, a nd so on. The absolute pressure in the c hamber needed to sto p the reacting process completely in this case is the injection pressure, \'ih ich in pra ctice is about 200 pounds pe r squa re inch abov e the normal chamber pressw·e of 300 pounds .}'.Br s qu a r e inch a bsolute. The period of t he oscillations v10uld de pend on the lengt h of the fuel lines, but t h e observed period of 0. 2 to 0.1 s e cond, ap prox i ITa tely, is not an unre a sona ble value to expect for this sort of oscill a tion. More deta iled considera t io n has bee n g iven to this problem by an other section of this project. The question of whether or not thi s inst ability enters in could be decided by u s ing a very high i n ject ion pressw·e with a l arge amount of resistance in the rea ctant lin es. This ha s not been done o n a l a r ge scale. - 27 - The second type of inst abili ty t ha t may enter prob a bly has a 2 very high fre-q_nency, of the ora_er of 3 x 10 or higher. It may be possible t hat t hi s oscill a tion couples with the first type to g ive the observed fre quency. The reason advanced fo r this possible inst a bility is that, assuming vaporiz a tion to be es s ential for a reasonably complete re a ction, an increase in pressure of only 105 pounds per s qu ar e inch c an completely reverse the evaporation from the nitric a cid in tl::e droplets, if it occurs befor e the tempera ture of the droplet has chant;ed appreciably. This v a lue is estimated from t he value s of the bubble p oint s for red and white fu ming nitric a cids g iven in Appendix B. At 300 pound s pe r s quare i nch abs olute, it is estimated tha t t he dew-point temperature of nitric a cid contai ning 0.15 weight fra ction nitrogen dioxide is 289° F. At this temperature tbe bubble-point p ressure is e stimated to be 435 pounds per s quare inch a bsolute. The increa.sed pressure re quired to stop the evaporation is very much less than this for acid th a t contains le s s nitrogen dioxide. Also it is probable that the difference between bubble and. dew points for gaso:J_ine at a g i ven temperature is larger t han for the red fuming acid. For one sample investi gated by Bahlke and Ka,y (10 ), at the temper a ture of 560° F, corresponding to a dew point of 300 pounds per square i nch, the bubble-point pressure is 510 pounds per square inch. If such an interru p t i on of evap oration occurred it would have to occur in a fairly short time before the temper a ture of the drop increa sed markedly. An estimation of this possibility can be ha d. Consider tha t tre injection velocity is a bout 100 feet per second and tbat t h e drop prob a bly tr avels one-tlri.rd of the chamber leng th of 1 foot before evaporation. This would - 28 - indic a te that the pressure increase would have to take pl ace in less than 3 2 3 x 10- seconds, or with a frequency of a t least 3 x 10 pe r second. The velocity of sound in the chamber is about 3140 feet per second for t he example of full-scale e quipment given earlier. Thus a pressure wave could he tra nsmitted to a ny part of the chamber in 3 x 16'1 seconds. After the slowil16 down of evaporation due to a slight increase in pressure, t lhe pressure would drop because the reaction had been slowed, the temperature of the acid droplets in t he hot chamber would rise somewhat, ancl finally a flush of gas would 0<ecur, resulting in a l a r ger mas~. of reacting materi al and therefore a rise in pressur e. Thus oscillation could be set up. It is possible tha t su}X')rsaturation of tbe droplets could aggravate the situa tion. The process is possible if these assumptions are true. Wo pulses of this freque ncy have been mea sured. However , it would. be recommended t hat an attempt be made to observe t hem if further work is to b e done on the use of t his reaction withou t preheatinge The indicated procedure to cure either type of inst ability is to prehe at the re actants. If tbe first tyr,e is predominant, prehea ting will decrease the ti me required for the re action after the reactants enter the chamber. If the second. type is important, preheating will eliminate alt ogether uncertainties in t he process of vaporization. In the case of the aniline-acid rea ction discussed in the second part of this dissertation, the second type of instability presumably does not enter, because it would take extremely large pressures to affect a reaction of the li quid-phase type, even assuming tbat pressure influences it adverse ly. It can also be concluded that gas analysis is of definite value in - 29 - estimating temperature and µJrformance of the jet unit. - 30 - PART II. THE REACTION OF NITRIC ACID A..1-JD ANILINE It was found by others (11) of this project tbat the reaction of nitric acid and aniline was well behaved a;nd stable on the l arge sca le. The limitati ons were efficiency of the weight ~a tio usoo.(1.5 to 1) and the high freezing point of an iline (21. m~o F). It was desired to study the r e action to ascertain if improvements could be made. A. Crucible Tests with Nitric Acid and An~li~ Purpose: In order to get the feel of the reaction and to a scertain the effect of a dditive agents in the aniline for freezing point lowering u p on t h e spontaneity , it was decided to investigate the r eact ion of t he re a ctan t s mixed relatively slowly in a small crucible. APPa ratus and Procedure: A porcela in crucible app roxi ma tely 1 inch in diameter, l inch deep, and 0 . 03 inch thick, insulated from its sur r oundings by an a ir jacket, was used. OoOOl pound ( 0.4 gra.rn ). ':!:he total amoun t of re a ct nnts used was about Doubling t h is amount did not a.ppreci ably affect the ti me re quired for reaction. A chrome l -alwnel thermocouple using wire 0.010 inch in diameter was pl a c e d i n the bottom of the vessel. One re acta;n t, usually the aniline , b ecause the a cid had a corro sive effect on the t hermocouple, was pl aced in the crucfole, bavi ng been me asured volumetrically fro m a rnicropipette . Then the other re a ctant was introd.uced in as sh ort an i nterv21 a s po ssible from anothe r p ipet te kept a t the s a..rne temperature as t he c ruc i ble. The temperature of t he reactants was observed on a critically damped galV MO~eter of a period of 0 .5 second connected to a potentiometer. I n ge neral the temperature of t he re a ct ants would rise at a - 31 - fairly slow rate, of the order of 100° F per second, to about 3700 F, after which temperature fla,'ile wa.s observed and the rate of i ncrease of temperatm·e was too rapid to be followed with the apparatus used. The time interval beginning with the introduction of the second reactant and endi ng at the ignition was measured. Results: It is emphas i zed that these exp:ni1rents do not exhibit the reaction under the same conditions as it occurs in jet-propulsion equipment. 'l1he time • re quired for complete mixing is at least two orders of magnitude larger in the present experiment, and the heat losses are greater. The former, however, is of much greater importance, as is borne out by the rate experiment. The a c id used had a speci f ic weight of 97. 781 pounds per cul)iC foot at 77° F, corresponding to 0.155 weight fraction nitrogen dioxide if no water is present. The aniline used was of c.p. grade with a boiling range of 361.4° to 366 .8° jf ancl with less than 0.0004 per cent non-volatile material present. Figure I* shows the time delay associated with the ignition of nitric acid and aniline at 77° Fas a function of mixture ratio. Each point is an average of 5 to 10 experimental poj_nts, some of which may differ from the average by as much as 50 rer cent of the average value.. This difference is probable due to the fact that much of the mixing was fr om convection from the heat generated in the reaction. A minimum aver2ge time delay is obtained in the vicinity of 3 OD 4 weight r a tio of a cid to aniline. Later equilibriu.11 calculations and experiment on a l a.rge scale have indic ated a maximum of - - - - - - - - - - - - - - - - -- - - - - - - * Figure s indicated by roman numerals refer to data. arabic numerals contain descriptions of apparatus .. Figures referred to by - 32 - :reaction temperature in this region, but it is clear that such conclusions canno t be extended to lar ge scale from the present experiment. Appendix A contains data relating to the efficient freezing point lowering substances for aniline, acetone, a nd methanol. Having ascerta ined the crucible t e st behaviors of a mtxture ( acid arrl aniline) whose behavior in jet - propulsive apparatus was known, it was desirable to try crucible tests on unknown mixtures, such as acid and aniline cont a ining a cetone or me thanol. Thi s was done, . and the results are shovm in Fi 6-ures II and III for a cetone and methanol, respectively. Compositions corre sponding to po ints e nclosed by the broken li nes are within the region judged to be s a tisfactory with respect to time of i g nitio n, compared to t he nitric a cid-aniline rea cti on. It is seen t h a t i n t hi s experimant it is possible to add to the aniline up to about 0.25 weight fraction acetone with out impairing i g nition. is lower for me thanol, pe rhaps 0.15 we i ght fraction. This fi gur e Lowered temper a tures did not greatly influence results. The conclusion is t hat, until da ta from actua l lar g e-ncale runs are available, acetone is to be "t$ ed instead of me thanol because it has a less deleterious effect on the react ion in the crucible than has me thanol. B. Ra te of Reaction Experiment. Pur·oo se: It was l101)ed to desi gn an apparatus t hat wruld g ive the time- temp er~ture history of the reaction of nitric a cid and aniline. It was pro :p osed to force the re acting ma terials a t a known rapid rate thr ough a tubi ng and to mea sure the temperature a t different points of t he path. re a ction prov ed to be rapid enough to render the obtaining of such data The - 33 - impractical. However, useful information was obtained from the exp:iriment. Apparatus: The system used to supply the reactants a t a known rate, pressure, and temperature was exactly that described ee.rlier and shown in Fi&:ures 7 and 8. The rate apparatus itself is shown in Figures 14 alild 15. Referring to Figure 14, the red fuming nitric ac id entered through the fitting ,A, passed through the check valve ::§., the nozzle TI, and into the chamber .Q.. Aniline was forced into .Q. through a similar fitting ,h,J. The reacting mixture passed downwa rd thrcugh a tube~, on which thermocouples were located at predetermined distances. A l oose-fitting st a inless-steel cylind.er 1: occupied most of the spc1.ce above the two h orizontal holes leading into .Q.. A shear disc .§. was provided to permit the pl:ig P to be blown out if high pressures were encountered. All pe.rts were of stainless steel with the exception of the gaskets and the shear disc, which were of lead. Procedure: In actual operation, a liquid such as kerosene, that does not react spontaneously with nitric acid and is relatively insoluble in aniline, was intro duced into a part of aniline line next to the appa ratus. This enabled the f low to be started a nd adjusted before the nitri c acid and aniline c 2 _me into contact. A mixture r at io of '"f, much leaner than the optimum, which lies between 3 a.nd 4, indic ated by crucible tests a nd equilibrium c e.lcula t ~ons, was chosen for the fir s t trial. fraction nitro gen dioxide. The acid used contained about 0.,15 weight The tbermocouple irrmediately below the chamber was observed continuously, and pr ovision wa s made to re a,d the other six in rapid succession. - 34 - Results: The consequence of this experiment was a rupture of the shear disc i of Figure 14 and of the tubing ~ 9 just below the fitting. The time required for the reactants to pass through the chamber Q. and the 2.5 inches of narrow tubing to the point~ was 0.037 secondo The temperature of a point on the tubing located 6 inches below~ did not cbange as much as 50 F during the operation. Thus the time required for fairly complete reaction was pro- bably about 0.03 second. When compared to the previous experiment, this shows that the time required for ignition of this reacti on depends primarily on the mixing of the reactants and not on the conditions of heat transfer to the surrou.~dings. In the present experiment, the heat transfer from the reactants to the surroundings was very good, the reactants being in a small cavity, 0.0084 cubic inch in volume (average linear dimension about 0.2 inch), in a large steel block at room temperature. The time for fairly complete mixing was al8o very short, probably of the order of 10~2 seconds, the velocity of the fluid in the nozzle tip li being about 50 feet per second, in the chamber~ about 2 feet per second, and in the fitting and tubing after the chamber about 20 fee t per second. The crucible experiment described previously gave reaction times of the order of 3 seconds, with the heat transfer to the surroundings per unit volume of reactant lower by at least a factor of 3, but with mixing time of the order of 1 second. Thus it can be concluded that mixing and not heat transfer plays the main role in time r~quired for ignition, for mixing times of the order of greater than or equal to 10-2 seconds, and that the reaction is therefore of the liquid-phase type. - C. 35 .. The Analysis of the Products of Reaction of Nitric Acid and Aniline. Purpose: It was desired to ascertain experimentally the products of the reaction of nitric acid and aniline in order to evaluate efficiency and reaction temperature and to check the values calculated assuming equilibrium to see if equilibrium actually obtained. Apparatus: The system used to supply the reactants is the same as that used earlier. The reaction ch8mber 9 the same as that used in the nitric acid and gasoline ignition experiments with injection jets intersecting at 90 degrees, is shown in Figures 16 and 17 • A slight modification was made in the valve seat to permit the eroded parts to be renewed when necessary. Molybdenum or copper inserts were used. The sampling apparatus was similar to that used earlier and will not be described here. The Orsat gas-analysis apparatus was the same as described earlier, but since no nitrogen dioxide was found in the products of any reaction studied, a more exact gas analysis procedure was possiblee Solutions used were potassium hydroxide for carbon dioxide, ferrous sulfate for nitric oxide, alkaline pyrogallol for oxygen, fuming sulfuric acid for unsaturated hydrocarbons, followed by copper oxide at 500° F for water and carbon monoxide, and a combustion pipette for hydrocarbons. The water present w ;,s ascertained gravimetrically from ice-cooled traj)s in the sampling appare.tu.s 9 but the final values shown in the tables are estimated from material balancee In one instance the carbonaceous residue in the sample line and traps was analyzed and measured. The value for the weight fraction nitrogen dioxide present in the nitric acid computed from specific weight measurement may be a few :i;sr cent difference too low, because of the presence of water probably in amount of not more than 3 per centG rre asurement s were made ... This was not known at the time the - 36 - Procedure: The reactants were forced into the chamber where the reaction started spontaneously. The regulating valve was adjusted to yield the desired pressure» and after allowing about 10 seconds for steady-state conditions, the sample was withdravm. over a period of about 50 seconds. Results: Four ru.ns were made with nitric acid and aniline at various pressures from 14.7 to 515 pounds per square inch. in Table VIII. Table IX. The results are shown The results of equilibrium calculat i ons (12) are shown in It is seen that agreement of the various components is fairly good in the case of pressures of 315 pounds per square inch or higher. For lower pressures it is seen that the equilibrium between nitric oxide, nitrogm, and oxygen is not satisfied, as was found for the atmospheric pressure reaction of acid and ga,soline.. As a matter of interest, the calculated reaction tem- peratures are plotted in Figure IV with the reaction temperature calculated for the two runs of 315 and 615 pounds per square inch of pressure. Specific impulse, assuming isentropic expansion to atmospheric preesure, is also indicated. The theoretical highest adiabatic reaction temperature is about 5100° F, which occurs at about a weight ratio of 3 to 1 of acid to aniline .. Experiments carried out by other sections of this project with large-scale apparatus indicate a maximum between 3 and 4. The effect of pressure on temperature is small, being not larger than 0.4° F per pound per square inch .. I'he result of the calculated composition :is shown in Figure V. At the weight ratio of 3, the principal products are water, carbon monoxide, nitrogen, and ~arbon dioxide, with some hydrogen and traces of hydroxyl, atomic hydrogen, ,xygen, nitric oxide, and. atomic oxygen.. It is expected that mixture ratios sreater than 3 would give very corrosive products because of the presence of - 37 - molecular and atomic oxygen. In general, on the r ich side of the mixture ratio 3, whe re there is a scarcity of oxidizer, the water-gas equ.ilibritim determines to a large degree how much energy will be released by the reaction. On the lean side of the mixture ratio 3, the reaction between carbon monoxide, oxygen, and carbon dioxide is most important. The calculations which lead to the composi tions t abulated assume t:t;ie following: 1. Equi librium is actually attained. 2. The fugacity of the gaseous component h in the products is the total pressure times t he mole fraction of h present. 3o The gas obeys the equation of state for a perfect gas; that is, the gas is thermally perfect. 4. The hea.t capacities of the monatomic gases are taken as half the values for the corresponding diatomic gases. 5. The react ions considered are: CO2 + H2 = CO + ¾O 20 = 02 1 OH + ~2 = H0 2 1 Hz + ~2 = H20 1 NO + ~2 = N0 2 2N = N2 NO = ~2 + ~2 2H = ¾ 02 + 0 = 03 , C(graphit e) + ~2 = co 6. 1 CO + iPz = CO 2 1 The laws of perfect solutions apply to the nitric acidni t rogen dioxide system used as an oxidizer. The starting point used in the calctua tion of an equilibrium composition and temperature f or a given composition of reactants was the assumption of a reacti on temperature. A se t of simultaneous equationg involving (i) the equilibrium constants evaluat ed a t the assumed t emperature, (ii) the material balances of the reactions considered, and (iii ) the given - 38 - chamber pressure was solved to give an actual possible composition of the products. This composition was the true one if the relation between en- thalpy change, weight fractions and specific heats of the products, and assumed reaction temperature stated in Part I-B was satisfied. If this relation was not satisfied, another temperature was assumed, and the tru.e result arrived at by trial and error. The exhaust velocity the gas would have in a jet unit was computed from the overall energy balance, assuming isentropic expansion for a calorically perfect gas to atmospheric pressure. Specific thrust was computed from Newton's second law. These computations were carried out by other members of the Department of Chemical Engineering. Similar experimental data with acetone in the aniline for the purpose of freezing point lowering of anil ine are shown in Table X. No difficulty was experienced in the ignition or stability of the reaction at any time, using approximately 0.2 weight fraction acetone in the aniline. Calculated data for this reaction are presented in Table XI, and comparison is made between calcu.lated and experimentaly evaluated temperatures in Figure VI. is seen that the agreement is good f or press~res above 300 Again it pounds per square inch. Ortho toluidine, whose freezing point is lower than that of aniline, both of which are discussed Appendix A, was tried next. factorily in every respect in this apparatus. It operated satis- The experimental data are in Table XII, the calcu.latei in Table XIII, and the comparisons between experiment and calculations in Figure VII. Orthotoluidine and aniline fonn their lowest entectic of -26° Fat 0.81 ortbotoluid.ine in anilinQ. discussed in Appendix A. This is completely Accordingly this mixture was tried with red fuming - 39 - nitric acid. The experimental results and calculated values are in Tables XIV and XV, respectively. The comparison is in Figure VIII9 Orthotoluidine is more desirable tban acetone as a freezing point lowering agent because orthotoluidine has a very low vapor pressure and hence a negligible inflammability hazard with air. D .. Summary. The conclusions to be drawn are as follows: The aniline-red fuming nitric acid reaction is of the liqu.id-pbase type for the mixing times investigated, since the rapidity of the reaction depends on the mixing in the liqu.id phase. That is, mixing is of controlling importance, rather than heat loss, in the time required for ignition, insofar as mixing times of the order gre·a;ter, than 10-2 seconds are concerned., The reaction can be predicted with accuracy satisfactory for present engineering purposes by means of e4Uilibrium calculations, provided that sufficiently high pressures are used. Other fuels that operate satisfactorily . with- red fuming nitric acid at ordinary temperatures are aniline containing up to 0.2 9 probably up to 0.3. weight fraction acetone, ortbotoluidine, and a mixture of 0.,19 weight fraction aniline in ortbotoluidine. - 40 - ACKNOWLEDGMENTS Sincere appreciation is expressed to Professor W. N& Lacey and Professor B. H. Sage of the Department of Chemical Engineering, and to Professor T. von Karman, Professor c. B. Millikan, and Doctor F. J. Malina of the Guggenheim Aeronautical Laboratory for suggesting these problems. The assistance of certain members of tr~ Department of Chemical Engineering was invaluable. Acknowledgment is made to M~. J.M. Green for assistance with tbe experiments and for development of certain analytical techniques, and to Mr. R. N. ·,'Wimpress for calculation of the equilibrium compositions. The copper reaction chamber and rate of reaction apparatus were designed in part by Mr. Glen W. Miller. The cooperation of certain personnel of other sections of the project, including Doctor M. Summerfield, was the essence of one experiment and helped immeasurably in others. Bibliogra.pby 1. International Critical Tables 3 £• 59. 2. Klemenc and Rupp, Zeit. fur Anorg. Chem., 194, 55. 3. Summerfield and Forman, Progress Report on the Development of a Liquid Propellant Jet Motor, Guggenheim Aeronautical Laboratory, California Institute of Technology, Project Noa, Report No. 10, September 25, 1941. 4. Wa lker, Lewis, McAdams, and Gilliland, Principles of Chemical Engineeringe 't hird edition, p. 112, McGraw-lI1ll Book Company, New York (1937) o 5. Bichowsky and Rossini, Thermo Chemistry of Chemical Substances, Reinhold Pub. Corp., New York (1936). 6. Guggenheim Aeronautieal Laboratory, Project No. 1, Heport Noa.11 (Chemical Section), California Institute of Technology, March 15, 1942., Values of heat capacity were taken from the following articles: a. b. c. d. e. f. g. h. Johnston and Davis, Jour. Am. Chem. Soc., 56, 273 (1934). Johnston and Walker, Ibid, 55 9 197 (1933); 57 9 683 (1935). Davis and Johnston, Ibid, 2§., 1045 (1934). Kassel 9 Ibid,.§.§., 1841 (1934). Gordon, ,J 0 ur • .Chem. Phys., !, 69 and 54~ {1934). Whitmer, Jour • .Am. Cnem. Soc.,.§§., 2229 (1934)0 Johnston and Chapman, Ibid, 56 9 153 (1933). Heck, Mech. Eng., 62, 9 (1940). 7. Lewis and Von Elbe• Jonr • .Amo Chem,. Soc.• fil, 612 (1935) o 8. Markham and Kobe, Jonr • .Am. Chem. Soc., 63, 1165 (1941). 9o Stodola and Loewenstein, Steam and Gas Turb1Res, Vol~ I, p$ 135 9 McGraw-Hill Book Company , New York (1927). 10. Bahlke and Kay, Ind. Eng. Chem., _g±, 293 (1932). 11. Guggenheim Aeronautical Laboratory 9 Project No,. 1 9 Report No. 12 ( . ll' J .Malina) 9 Take-off and Flight Performance o:t" An A-20- A Airplane as Affected by Auxiliary Propulsion Supplied by Liquid Propellant Jet Units 9 California Institute of Technology, June 30, 1942. 0 12. Guggenheim Aeronatucial Laboratory, Project Noe 1, Report No. 14 (Chemical Section), California Institute of Technology, April 25, 1943 (Appendix A). L G I . II D C II w :x: V y F H Schematic Dr awi ng of Appar a tus for the Premixed Reaction of Nitri c Acid and Ga soline. FIGURE l R s N. p T Fig,fro 2 De tails of Mixing Cha.m'Jer for Liq:uid-Phase Reaction Equipment I I ~ Figure 3. Gas-phase Reaction. Apparatus for Nitric Acid and Gasoline. Jigure 4. Equipment Used for Sampling the Products of Reaction. s .' \ L' R 1 \ l Nz l ! I __.___ \ To X _,... \ , __ / i ', · A_} ~ ~ B I- r- '--- ' - - - - - - ''! -- ~ =h. K ~ I'---- Fir;ure 5 " ½ _J r '--' \ ) . I, 11 - , ' J ' --- - _J . _j_ -- C Throttle F ~--- C -- -1 ,I ___-- ---\- Jr- -- G: I: [l_.=-- ---_, H V 'L ·u I ' • _---; -:::_)--, • ->~ --~ - __A _-- -- , \____' ' I ,..- I / 'I.! I ~~T "~· Ill;I L;-;; I I /°)( u,, , 1,l p Arren.;e,.ent of .8quip1:1ent for Gas- phuse r\ea.ction Investigation --, ' 0 I ' ,,,--, ~ ' rI I l •, •-+~.x--, - i I 11 i I l_J l: nn li I! x ~l i u T ___ I .J.__ I .✓ r- ·, l_C.r l I ' I J i.) I, •I I I I I, \~·- 1 I 1, r·~ - \._) ,-~~ I Figure 6. Gas Analysis Apparatus. ·-----·----- - - - - G , H' X o!'_ ~ ~C -'- ~o.....-t.--D-1 - _[ /\ --0 1• B' ,. L. J• • s Figure 7 Flow Sheet of Apparatus for Ignition of Nitric Acid. and Gasoline Using Opposed Injection Nozzles. _, ,, Figure 8. View of Equipment Used to Supply and Measure the Reactants. ---- 11 Figure 9. View of Heat Transfer Equipment. ' / ,:_ t/ ,;.. . I , ,....,. I ,_ , ~~ ,....,,,..-~-~-==-=.,__,.,,_.....,.,..._ ...._- - ' - · ·-- f:il ,< .,,....._~_ . . .. ~-.ll,,. / -;;:---1.-~. / :x..._...... ~~ .... L. --'-("'- \ ' /" / ,. ,,. ~ , r· . ~ '~~ /' .• • ';t-=tl L_, . / p·dlj -~-~---r--,,..----::,-~3-7 --::===-~------~. -~~~ i ~ G) ... rn ~ s::l ,ta> a J-4 E-t ~ :1 1%: Pi g. 0 M 'M "" .i ~ ...c ~ i::: G> § ~ /,I I ;;'/! - -- - - - - - - - - -- - - - - - - . / .&. t 1 ;r · Il i I , 1 1 Section 1-1 ~ ~- ' ,. J I I; l•,,~ tjl-j.-1'" ~ ? I. c J!11W . F JI :; ~lj ~ '~:1 ~I i r I I t~ )'4'U.re 11 .&rrangement of Reaction Chamber and Pres su.re Control Equipment u - ___ J Figure 12. View of Ignition Chamber Usi ng Opposed Jets. • G - -- ____ ...__ b 0 Fi~iire 13 . .Arrangement of Ignition Equipnent in Reaction Ch.a.mb"'r -------- -- /, ; , • / , / .{~ Figure 14. J.. I . Sketch of Bate of Reaction Equipment . Figure 15. ' f , l . l 1 L" ' I :6 ~ I !{ l f :9 5 1, Bate of Reaction Equipment. l l 1 ll 1 0 1 2 Figure 17. ~ View of Continuous Reaction Chamber and Control Valve. ( TABLE I. SUMMARY OF RESULTS FOR MINIMUM PREHEATING TEMPERA.TURES FOR THE GAS-PHASE REACTION OF NITRIC ACID AND GASOLINE AT ATMOSPHERIC PRESSURE Ni t ric Acid Sample Number 209 214 -- Weight Fraction Nitrogen Dioxide from Density (2) 213 0.17 Weight Fraction Water from Density (1) 0.465 0.156 Weight Rate of Flow, lb/sec 1 to 6 X 10-4 4.4 X 10-4 4.2 X 10-4 Weight Fracti.on in Mixture o.67 to 0.92 0.844 0.875 0.8650 0.8417 Theoretical Weight Fraction Assuming Nitrogen as Nz Theoretical Weight Fraction Assuming Nitrogen as NO - 0.9129 0.9014 Preheating Temperature. 0 r 750 700 490 Sample Number 210 210 210 Carbon/Hydrogen Ratio, lb/lb 5.884 5.884 5.884 Weight Rate of Flow, lb/sec 0.5 X 10-4 0.83 X 10-4 0e55 X 10-4 Weight Fraction in Mixture o.33 to o.oa 0.156 0.125 Preheating Temperat are, °F 750 460 490 Gasoline Flame Volume (visual estimate) cu.ft /lb/sec No reaction TABLE II. COMPOSITION OF THE PRODUCTS OF REACTION OF RED FUMIN~ NITRIC ACID AND GASOLINE Atmospheric Pressure Component 0.120 Weight Fraction Gasoline 0.159 0.266 Nitrogen and Hydrogen 0.1483a 0.1430 0.3430 Nitrogen Dioxideb 0.1685 0.0442 0.0000 Nitric Oxide 0.0000 0.1410 0.0545 Oxygen 0.0000 0.0000 0.0000 Carbon Dioxide 0.2150 0.2400 0.1195 Carbon Monoxide 0.0000 0.0000 0.1730 Water 0.4381 0.4052 0.,3095 Nitric Acid 0.0301 0.0266 0.0005 a Compositions are expressed as mole fractions. b The measurements at 0.266 and 0.120 weight fraction gasoline involved only a measurement of the total qumtity of the oxides of nitrogen. The absence of nitrogen dioxide at the higher weight fraction of gasoline was estimated from the colorless nature of the gas while the absence of nit ri c oxide at the lower weight fraction was indicated by the material balance and by the lack of change of color upon the admission of oxygen. TA:BLE III. THE REACTION OF NITRIC ACID AND GASOLINE AT ATMOSPHERIC PRESSURE Component 0.120 Wei;!dlt Fraction Gasoline 0.159 o.266 Reactants 5.965a 1.405 1.000 4.265 1.005 1.000 2.2094 0.5(304 1.000 Nitric Oxideb Nitrogen Dioxideb Nitrogen Carbon Monoxide Carbon Dioxide Carbon (free) Hydrogen (free) Water Nitric Acid Total Weight 0.000 2.700 0.969 0.000 2.160 0.248 0.000 1.849 0.444 8.370 0.876 o.420 o.730 0.000 2.140 0.259 0.000 1.505 0.340 6.270 0.268 0.000 0.796 0.865 0.277 0.075 o.916 0.02s 3.730 Enthalpy Change or Reactionc 15?0.3 1912 1896 Reaction Temperature, °F 4060 4160 3780 Nitric Acid Nitrogen Dioxide Gasoline Products o.505 a All weights are expressed in pounds of reactant or product per pound of gasoline. b Only total quantity of oxides of nitrogen was measured at weight fraction of 0.120 and o.266 gasoline. c Enthalpy of reaction is expressed in Btu per pound of reactants for both reactants and products at 64° F and 1 atmosphere pressure, the product water being considered in the liquid state. TABLE IV. COMPOSITION OF PRODUCTS OF REACTION OF NITRIC ACID AND GASOLINE WI 1.I1H FULL-SCALE JET-PROPULSIV&J EQUIPMENT Component Mole Fraction Nitrogen Dioxide 0.0625 Nitrogen Tetroxide o.6475 Carbon Dioxide 0.1619 Oxygen 0.0026 Hydrogen 0.0223 Carbon Monoxide 0.0335 Nitrogen 0.1216 Water o.4613 Nitric Acid 0.0023 Free Carbon 0.0845 Operating Conditionsa Reaction Temperature, °F 3880 Exhaust Temperature, °Fb 2175 Enthalpy Change in Nozzle , B\u/lb 725 Exit Velocity, ft/sec 6014 Specific Thruot, lb.sec/lb 188 Weight Fraction Gasoline in Reactants 0.141 a T'nese conditiona were computed from analysis of the products of reaction. b Calculated for isentropic expansion from 300 pounds per square inch to 14.7 pounds per square inch absolute pressure. 6.3 6.3 6.3 16.6 16.6 16.6 75 75 75 75 Pressure Acid Jet Above Chamber (lb/in.2) 15 0 100 9.1 1 5 15 0 9.1 9 15 0 100 12 16 0 Ignition No. of Power Trials (watts) 9.1 6.4 mosphere) Weight Chamber Pressure Ratio Acid/Ga- (lb/in.2 soline above at- 100 100 Pressure Gasoline Jet Above Chamber (lb/in.2) a Nitric acid used contained 0.19 weight fraction N0 2 , from specific weight measuremen1E. Reactants at 77° F. 6.3 14 Acid Jet Gasoline Diameter Jet Diam. ~in.x 10-3) (in.x 10-3) no yes no yes Use of Air /. 0 1 0 :6 Ignition TABLE V. THE SPARK IGNITION OF NITRIC ACID AND GASOLINE WITH OPPOSED JETSa 1 4 9 6 No Ignition 4.48 250 250 250 200 200 200 200 150-250 390 20 -- 10 -- 4.48 -- 250 250 9. 1 1 1 6 1 1 2 14 3 a Nitric acid used contained 0.19 weight fraction N0 from specific weight measurerre nts. 2 -- 390 20 20 4.48 390 390 390 20 20 20 25 25 0 4.48 4.48 250 200 4.1 7.5 390 20 25 5.01 250 250 4.1 7.5 392 20 0 9.2 100 75 6.3 16 .. 8 397 100 0 9.1 100 75 6.3 16.6 380 60 0 9.1 100 75 I gnition Surface No. of Power Density Trials (watts) af I gnition Power (watt s/ in. 2) Chamber Pressure (lb/ in.2 above atmosphere) Weight Ra tio Acid/ Gasoline Pressure Gasoline Jet Above Chamber (lb/in.2) Pressure Acid Jet Above Chamber (lb/in. 2 ) 6.3 Gasoline Jet Diam. (in.:x: 10-3) 0 1 1 1 2 1 1 2 14 3 I gni tion 9 0 0 0 4 0 0 0 0 0 No I gnition Reactants at 77° F. -- no no no no no no no no no Use of Air TEE HOT WIRE IGNITION OF NITRIC ACID AND GASOLI:NE WITH OPPOSED JETSa 16.6 Acid Jet Diameter (in.x 10-3) TABLE VI. 150 150 200 200 5.4 5 .. 4 5 .,4 5.4 5.4 5.4 5.4 5. 4 5.4 5.4 5.4 9.0 ~.o 9 .o 9.0 9.0 9.0 9.0 9.0 9.0 9.0 9.0 250 250 250 250 250 250 250 350 150 100 150 200 Pressure Gasoline Jet Above Chambe~ ( l b/in. ) 0 0 0 0 0 0 0 0 0 0 0 0 mosphere) Chamber Pressure (lb/in.2 above at- 3.70 3.70 3.70 3.70 3.70 3.20 3.30 80 80 80 89 80 80 80 80 80 -2.70 80 80 80 Ignition Power (watts) 5.83 4.75 4.12 Weight Ratio Acid/Gasoline 400 400 400 400 400 400 400 400 400 400 400 400 Surface Density of Power (watts/in.2) a Ni t ri c acid contained 0.137 weight fractio n nitrogen dioxide, from specific weight measurement. 200 200 200 150 200 200 200 200 5.4 9.0 Pressure Acid Je t Above Chamber (lb/in.2 ) Gasoline Jet Diam. (in.x 10-3) THE HOT WIRE IGNITION OF NITRIC ACID AND GASOLINE WITH JETS I:NTERSECT!NG AT 90 DEGBE}Ea Acid J et Di ameter,,. (in.x 10-0 ) TA:BLE VII. 3.70 3.72 3.72 0 0 0 0 0 300b 250 250 250 250 350 350 200 200 200 200 300 300 5.4 5.4 5.4 5.4 5.9 5.9 9.0 9.0 9.0 9.0 9.7 9.7 b Unstable. 3.70 3.70 :5. 70 3.70 0 250 200 9.0 5.4 Weight Ratio Acid/Gaaoline Chamber Pressure (lb/in. 2 above atmo sphere) Gasoline Jet Diam 3 (in.x 10-) Acid Jet Diameter (in.x 10-3) Pressure Gasoline Jet Above Chamber (lb/in.2) (cont 1 d) 80 80 80 80 80 80 80 Ignition Power (watts) 400 400 400 400 400 400 400 Surface Density of Power (watts/in.2) THE HOT WIRE IGNITION OF NITRIC ACID AND GASOLINE WITH JETS INTERSECTING AT 90 DEGREES Press1.tre Acid Jet Above Chamber (lb/in. 2 ) TABLE VII. TABLE VIII. COMPOSITION OF PRODUCTS OF REACTION OF ANILINE AND RED FUMI:NG NITRIC ACID Run Number 15a 15b 19 27 Weight Fraction Nitrogen Dioxide in Acid Phase 0.145 0.145 0.153 0.154 Weight Ratio Acid / Fuel by Orifice Measurement 1.57 1.57 2.37 1.89 Weight Ratio Acid/ Fuel by Material :Balance 1.93 1.93 2.10 1.83 14.7 115 315 515 0.2435 0.1830 o.0429 0.2321 0.2985 0.0230 0.3556 0.0734 0.2076 0.3404 0.1299 0.2158 0.4310 0.1949 o.02s4 0.0900 0.2385 0.4573 0.1889 0.0253 0.1237 0.0929 0.0218 0.1179 0.1517 0.0130 0.2011 0.0415 0.1173 0.1922 0.0783 o.2076 0.4920 0.4349 0.1096 0.1821 0.3637 0.1645 0.0240a 0.1561 3940 4800 4270 3857 5740 6841 6956 178.6 212.7 216.3 ') Reaction Pressure (lb/ in • ..,abs.) Dry Gas Analysis (mole fraction) Nitric Oxide Carbon Dioxide Hydrogen Carbon Monoxide Nitrogen Hydrocarbons Total Products Analysis (mole fraction) Nitric Oxide Carbon Dioxide Hydrogen Carbon Monoxide Nitrogen Hydrocarbons Water Carbon Calculated Reaction Temperature ( 0 1!,) Exit Velocities (ft/sec) Specific Impulse (lb.sec/lb) ~ Assumed equivalent :> Equivalent to CR4 to c1 • 54n5 • 07 -- 0 • .3980 0.16~ 0.0220 0.1297 3 2 1 Weight Ratio Acid/ Fuel 4930 5065 7091 220.5 0.1594 0.0013 0.0003 0.0550 0.4163 -- 3710 6514 202.6 0.0098 o.3466 0.1695 1950 5207 Reaction Tem~rature (°F) Exit Velocity (ft/sec) - Specific Impulse (lb.sec/lb) 161.9 --- o.3446 0.0173 -- -- 600 192.54 207.95 175.9 5658 6192 6688 o. 3422 0.1738 --0.0100 2176 6998 3710 239.2 7694 5150 0$0002 0.0626 o.2881 0.0092 0.0279 0.1622 0.2494 - 0.2126 0.1594 0.0013 0.0003 0.0550 0.4163 -- 0.3407 0.0211 0.1930 0.0032 -- 0.0027 0.0015 3 0.15 -- -- -- 2 0.15 0.1551 0.1122 --- 1 0.15 2015 -- 0.0016 0.2898 0.0003 0. 0338 0.2265 0.0000 0.2246 0.0163 0.0001 0.1963 0.0026 7 0.15 4320 -- -- 0.0002 0.0105 0.2919 0,.0038 Q.0622 0.2431 0.0601 -- 0.0002 o.0632 0.2827 0.0114 0.0316 0.1587 0.2510 0.1551 0.1122 ---0.2126 0.1919 0.0036 -- -- 0.2136 0.0171 0.0037 0.0020 Oxygen Monatomic Oxygen Triatomic Oxygen Nitrogen Nitric Oxide Nitrogen Dioxide Monatomic Nitrogen Hydrogen Water Mona.tomic Hydrogen Hydr oxyl Carbon Dioxide Carbon Monoxide Carbon -0.0898 o.0078 5 0.15 -- Composition of Equilibrium Mixture (mole fraction) 0.15 0.15 0.,15 300 AND RED FUMING NITRIC ACID .f!.!f EQUILIBRIUM CALCULATED COMPOSITION OF TEE PRODUCTS OF REACTION OF ANILINE Weight Fraction Nitrogen Dioxide in Acid Phase Pressure (lb/in.2) TABIE IX. T.ABLE X. COMPOSI'l1ION OF THE PRODUCTS OF REACTION OF ANILINE, ACETONE, .AND RED FUMING NITRIC ACID Run Number 38a 33 30 38b 34 Weight Fr action Nitrogen Dioxide in Acid Phase 0.2137 0.2100 0.1979 0.2137 0.2105 Weight Fraction Acetone in Fuel 0.1875 0.1923 0.2000 0.1875 0.1875 Weight Ratio Acid / Fuel by Orifice Method 2.05 2.. 04 2.01 1.86 Weight Ratio Acid / Fuel by Material Balance 1.76 1.88 1.76 2 Reaction Pressure (lb/in. abs.) 115 115 315 315 540 0.0020 0.1915 0.1750 0.1710 0.1405 0.1330 0.4515 0.0815c 0.2300 0.1190 0.4470d 0.0290 0.2275 0.1040 0.4505 9 o.0470 0.0018 0.1692 o.1568 0.1558 o.1242 0.1163 0.1175 0.3890 0.0720 0.2061 0.1039 0.1066 0.4006 0.0260 0.2073 0.0888 0.0948 0.4105 0.0428 1.79 Dry Gas Analysis (mole fraction) Oxygen Nitrogen Nitric Oxide Hydrogen Carbon Dioxide Carbon Monoxide HydrocarbonsCCnHm) 0.2410 0.0485 0.1580 0.1435 0.3730 0.0360a 0.2125 0.0703 0.1262 0.1680 0.3520b 0.0710 Total Product Analysis (mole fraction) Oxygen Nitrogen Nitric Oxide Hydrogen Water a Carbon Dioxide Carbon Monoxide Hydrocarbons Carbonf 0.1386 0.027 9 0.0909 o.2635 0.0825 0.2146 0.0207 0.1613 0.1423 0.0471 0.0845 0.2331 0.1125 0.2357 0.0475 0.0973 Calculated Reaction Temperature (°F) 4640 4395 3500 4035 3580 Exit Velocity (ft/sec)g 6082 5966 6048 6730 6762 185.5 188.0 209.3 210.3 Specific Impulse (lb.sec/ lb)g 189,.l Equivalent to C1.39H4.6l Equivalent to C1.21H3.65 Equivalent to C1.90H5.80 Equivalent to C1.35H4.07 e Equivalent to c1 • 57H5 .15 f Determined from material balance. C1 ° Calcula ted assuming isentropic expansion through jet to 14.70 pounds per square inch absolute. TA13LE XI. C.AICU-.LATF.D COMPOSITION OF TEE PRODUCTS OF REACTION OF ANILINE, ACETONE, AND RED FUMING NITRIC ACID .A!r EQ,UILIBRiu"M 2 Pressure (lb/ in. ) 500 500 500 Weight Fraction Nitrogen Dioxide in Acid Phase 0.15 0.15 0.15 Weight Fraction Acetone in Fu.el 0/. 10 0.10 0.10 Weight Ratio Acid / Fu.el 1 2 3 , Composition of Equilibrium Mixture (mole fraction) Oxygen Monatomic Oxygen Triatomic Oxygen Nitrogen Nitric Oxide Nitrogen Dioxide Monatomic Nitrogen Hydrogen Water Monatomic Hydrogen Hydroxyl Carbon Dioxide Carbon Monoxide Carbon 0.0031 0.0015 0.1081 0.1571 0.3446 0.0300 0.2053 0.1763 0.0010 0.0003 0.0600 0.4000 0.0595 0.1898 0.0034 0.0002 0.0154 0.3380 0.1639 0.3008 0.0090 0.0285 0.1682 0.2360 Reaction Temperature (°F) 1940 3780 5100 Exit Velocity (ft/sec) 5868 6998 7543 Specific Imp~lse (lb.sec/lb) 182.5 217.6 234.5 TABLE XII. COMPOSITION OF THE PROIUCTS OF REACTION 0RTHO-TOLUIDI1TE .A.ND RED FUMING NITRIC ACID Run Number 47 49 50 0.2182 0.,2163 0.2179 1.7 2.8 2.8 2.33 2.40 315 315 315 0.1630 0. 1639 o.4091 0.2146 0.1540 Q.,0020 0.1455 0.4420 0.2100 o.;;53° 0.0465° 0.1294 0.1304 0.1286 0.3077 0.1788 0.0161 0.033la 0.2063 0.1488 0.,3254 0.1707 0.1224 0.0015 0.1157 0.3513 0.1669 0.;;12b 0.2046 0.0370c 0.2052 Calculated Reaction Temperature (°F) 4325 4570 4440 Exit Velocitye (ft/sec) 6785 7017 6873 Specific Impulsee {lb.sec/lb) 210.7 218.2 213.7 Weight Fraction Nitrogen Dioxide in Acid Phase ,Veight Ratio Acid / Fuel by Orifice Measurement 1 Weight Ratio Acid / Fuel by Material :Balance Reaction Pressure (lb/in.2abs.) Dry Gas Analysis (mole fraction) Carbon Dioxide Oxygen Hydrogen Carbon Monoxide Nitrogen Nitric Oxia.e Hydrocarbons 0.1620 o. 3877 0.,2253 0.0203 0 . 0417a .-0.1871 Total Product Analysis (mole fraction) Carbon Dioxide Oxygen Hydrogen Carbon Monoxide Nitro gen Nitric Oxide Hydroa,arbons Water a Equivalent to cl.21H4.78 b Equivalent to cl.95H5.33 c Equivalent to cl.29H4.48 d Determined from material balance. e Calculated assu..ming isentropic expansion through jet to 14.70 pounds per square inch absolute. TABLE XIII. CALCULATED COMPOSITION OF THE PRODUCTS OF REACTION OF ORTHO- TOLUIDINE .Alm RED FUUING NITRIC ACID AT EQUILIBRIUM 2 Pressure (lb/in. ) 300 300 300 Weight Fraction Nitro gen Dioxide in Acid Phase 0.15 0.15 0.15 Weight Ratio Acid/ Fu.el l 2 3 Composition of Equilibrium Mixture (mole fraction) Oxygen Monatomic Oxygen Triatomic oxygen Nitrogen Nitric Oxide Nitrogen Dioxide Monatomic Nitrogen Hydrogen Water Monatomic Hydrogen Hydroxyl Carbon Dioxide Carbon Monoxide Carbon 0.0110 0. 3146 0.1884 Reaction Temperature (OF) 0.0012 0.0007 0.1026 0.1505 0.3506 0 .. 0328 0.2377 0.1520 0.0005 0.0001 0.0489 0. 4104 0.0001 0.0765 0 . 2972 000088 0.0208 0.1486 0.2593 1855 3490 4905 Exit Velocity (ft/sec) 5153 6452 7046 Specific Impulse (lb. sec/ lb) 160.19 200.62 219.l ' 0.1849 0.0019 TABLE XIV. COMPOSITION 0]' THE PRODUCTS OF REACTION OF ANILINE, ORTHO-TOLUIDililE, .AND RED FUMING NITRIC ACID Run Number 53 Weight Fraction Nitrogen Dioxide in Acid Phase 0.220 Weight Fraction Ortho-toluidine in Fuel 0.810 Weight Ratio Acid / Fuel by Orifice Measurement Weight Ratio Acid/ Fuel by Material Balance 2.23 Dry Gas Analysis (mole fraction) 0.1523 0.0015 0.1923 0.4195 0.0232 0.2112 Carbon Dioxide Oxygen Hydrogen Carbon Monoxide Hydrocarbons Nitrogen Tot al Product Analysis (mole fr action) Carbon Dioxide Oxygen Hydrogen Car.bon Monoxide Hydrocarbons Water Nitrogen Black Residuea 5 0.1214 0.0012 0.1532 0.3343 0.0184 0.2016 0.1683 0.00165 Calculated Reaction Temperature (°F) 4290 Exit Velocity (ft/sec) 6834 Specific Impulse (lb.sec/lb) 212.5 T.AJ3IE XV. CALCULATED COMPOSITION OF TF..E PROWCTS OF REACTION OF ORTHO-TOLUIDI NE, ANILINE, AND RED FUMING NITRIC AC ID AT EQ,U ILIBRIUM 2 Pressure (lb/ in. ) 300 300 300 Weight Fraction Nitrogen Dioxide in Acid Phase 0.15 0.15 0.15 Weight Fraction Aniline in Fuel 0.19 0.19 0.19 Weight Ratio Acid/ Fuel l 2 3 Composition of Equilibrium Mixture (mole fraction) Oxygen Monatomic Oxygen Triatomic 0,cygen Nitrogen Nitric Oxide Nitrogen Dioxide Monatomic Nitrogen Hydrogen Water Monatornic Hydrogen Hydr oxyl Carbon Dioxide Carbon Monoxide Carbon 0.0027 0.0017 0.1043 0.1523 o. 3481 0.2327 0.1533 0.0007 0.0002 0.0499 0.4109 0.0002 0.0740 0.2876 0.0121 0.0287 0.1450 0.2606 0.0314 0.0110 0.3187 0.1866 0.1846 0.0029 Reaction Temperature (OF) 1845 3540 5040 Exit Veloc ity (ft/sec) 5127 6476 7142 Specific Impulse (lb.sec/lb) 159.43 201.36 222.09 10 \ \ \ \ l, '\ '\ '' • I I \ I \ '\ ID ....El i-4 b__ r---..... 0 .5 0.6 o.7 I I I "'V 0.8 Wei ght Fraction Nitric Acid Figu.re I • Time Del.q Associated i'li th the . C.ruci ble Ign.i t ion of Ni trlc Acid and Aniline at 7P P. Figure lI Ignition of Aniline-Acatons-Hit:ri c Aci d Mixtm."e s Figures in parentheses are t eaoe1·atures of r eactgnts, °F . Figures p r eceding parenthe se~~ a r -2 i cni tion del ,,;y ; se con,ls , 0 0 Satisfactory I gnition Erratic I gnition (?) 0 .3 t--------+-------<..-~-t-------r------r--------, /\ I \ I __3__..\ :i (?) Oo 2 l-------+----,~c_._._!.,aJ;lfl--1,-L-;L------+--------ir-------1 \ I I \ \ I I \ \ I / \ 2.9 (7) \ 3.3 (?} 0.11------l---+--U_'--+-----c.T:,--;-..;._.,,--f-----~------i ~ I I / I / / I I 8 ( 18•~ •//; o.6 1 , 3 .18 ( I""-' ~o) - •o.7 - o.a 3o48 (681) Weight Fraction Nitric Acid 0.9 L,O I I I I Figtll'9 III Ignition ot .A.n111ne-Methanol-Nitrlo Acid Mixtur:e1 11go.res in parentheae; are temperatures of reaota:nte, °Fo Figures preceding those in parentheses are ignition delay expressed in second!. I I 0 Satiafa.ctory Ignition 6 Erratic Ignition O l~ .5 (68) I I\ \ ' \ I \ I 3.05 (68) - 3068 (141 ' 0.6 {14) I I I \ 8 (68 l \ ? I \ \ \ Oo5 \ \..3 .82 ( 60) I 2 .-13 ( 68 ) r,. o.7 I ,l 3.18 (p8) o.e Weight Fraction N1tric Acid LO 6000 / - --- - ~ / / ,. I -, I I I 5000 I I I ,I I / 4000 .:!R I I I Q) ~ fo 3000 ~ .. Ci r', E Q) E-< i:: 0 •M t) c.j (1) p:; 2000 ... _ -- -- -- -- - ... .... ... 220 .... ... - ~'' -......... eaction T~mperature ~ 200 ---............ r--, ~'- 180 b/ --0I 160 I 30 D pound.s p ~r square inch . -- ------ 6DO pounds per squar1 .inch ,'l ,D r-i 140 ,. . . ,c w 120 ,---{ c.: Figur e IV. Ad 1-abatic Re 11,ction Ter ~era.ture and S becific I trpulse Obt 11,ined fro1 Reactioi o f' d-n .line ,:mil Red Furnin I! Nitric . ,f"1rl 100 Assum nG Thenn 1>dynarriic E ~u.ilibriUJh u; r-i ~ p_ s H 0 •M Ct-i •r-1 0 Q ~~ (/) 80 E :iq erimenta value, 3 5 pounds per s_ql La.re inch 60 -0- E: ~erimenti 1 value, I~1s pound~ per s1 uare incl~ 40 1000 20 1 /J) ,-.. , (l) 0 G S--1 0 I -.µ 1-- - ~ / / .. ... r-------__ ----V,, - r---..---..._ ,// ' -- ---- - Spec fie Imp lllse 240 . ------- I ,' ,. ,. / 1-- 2 3 4 y; eLr;ht :fr,t lo Aci d/An il ino 5 6 7 /.0 0 .6 _ t ~ _..J__ _ _ _- + - - - - - - - + - - - - - - + - - - - - - + - - - - - + - - - - - ; 0./ \) ~ ~0.06---4-l~+~-..J.l-__:__ _~-----::::a,,.-+-,====,B~""""'==-----+------l ~ ~ a 04---l--l--+---/--l------+-~~---+--1-~--t--~----lr---""-..;;:-----i ~ 0 / .c J,1/E/GHT 6 -11- ..:J' RAT/0 7 6 AC/.0/FUEL Figure V. CALCVLATEO £QV/L/BR/i/M THE REACT/ON RE.O FU/\,t/NG CO/vfPOS/T/0/V PROOt/CTS N/TR/C AC/0. OF AN/L/Nc OF ANO 5000 / LL. I / ,~· 0 0 0 w 0-· 0:: :) I- C"::, 4000 ~~- <( 0:: pq w a.. ~ w I- 3000 v~o u ·. ~· ~ ~~ z 0 I- V 2000 <( w ~~ / 'vo 115 LB. PER SQ. IN. -<>-315 LB. PER SQ. IN. 540 LB. PER SQ. IN. 0 0:: ~ )000 ,' 0.5 1.0 1.5 2.0 WEIGHT RATIO ACID/FUEL f:\.t;'o..re V,L 2.5 ~::e.e_cti on '>.ni~:;e:r;:~tun :· of :, "i>': ,.x•:.: :e:' :.,.:i.:., i r. ·: ~atr ic: :,.c ic1 '.:,y;;t1:: ::'. o 3.0 ·, • ._,, •. _••.· •~ 6000r------i-------t-----,-----r----..--.----- 5000t----+----+----t----+----+---~ ', I 0001------1-------+--- o EXPERIMENTAL 1----- 3 I 5 LB. PER SQ. IN. 0.5 1.0 1.5 WEIGHT RATIO Figure VII. 2:0 ACID/FUEL 2-5 3-0 6000 5000 o ·· / u.. 0 a.-~ W' .. 4000 «-- 1- z ~ .v~ I 00 0 I- e:,O· , ~ 3000 <::)' u<( ..-<..~ ':v'?- w a: G0 . '?-v 2000 ,/ .. C, ., I 000 EXPERIMENTAL 315 LB. PER SQ. IN. 0 o. 5 1.0 1.5 WEIGHT .RATIO 2·0 2-5 ACID/ FUEL Influence of Mixture Ratio upon Flame Temperature in the Orthotoluidine-Aniline-Red Fuming Nitric Acid System Figure VIII. 3-0 Appendix A THE LOWERING OF THE FREEZING POINT OF ANILINE BY THE ADDITION OF CERTAIN COMPomms Because the freezing point of aniline, 21.02° F, is well above many temperatures encountered in practice, it is desirable to find the freezing poi•t lowering substance that is best from the points of view (i) of giving the best reaction with red fuming nitric acid and (ii) of giving a vapor pressure at ordinary temperatures when dissolved in aniline such that the presence of air at pressures of atmospheric or above will not introduce the hazard of inflammability. The apparatus used in determining the freezing points of different mixtures is depicted in Figure A-l. The sample is contained . in a small closed test tube A,, which contains a stirrer ;]! and a theri.nocouple Q.. Surrounding this test tube is a closed air jacket containing a drying agent, phosphorous pentoxide. The ou.ter test tube~ is immersed in a bath of acetone cooled by dry ice, which is contained in the transparent Dewar flask 12.• The copper-constantan thermocouple employed was calibrated against a Leeds Northrup Platinum Resistance Thennometer which in turn was calibrated by the Na.tional Bureau of Standards. It is believed that the uncertainty involved in the temperature of the sample is not more than 0.1° F. All tlE reagents employed were of c.p. quality except turpentine, ethyl cellosolve, and methyl-ethyl ketone. No attempts were made to purify any of these substances further, and it is believed therefore that the final results may involve uncertainties up to 2° F. Because of the pronounced supercooling that occurred, the measurement of the disappearance of the solid phase gave a much more reliable indication of the freezing point than did the reverse procedure. The freezing point was ascertained by the melting-point curve and checked by visual observation .. The results are plotted in Figures A-I, A-II, and A-III, along witl1 the values for acetone, which are available in the literature (1). Table A-I contains the numerical values. Acetone is the most efficient as a freezing point depressant, as can be seen by Figure A-I, but the fugacity of 0.3 weight fraction acetone in aniline is about 3.95 pounds-i;er square inch at 130° F. This high value of fugacity gives rise to inflammable mixtures with air present, if the total pressure is between 30 and 150 pounds per square inch. Appendix C. This is discussed in Similarly methanol and methyl-ethyl ketone are efficient but have high enough fugacities, when present in aniline to an appreciable extent, to constitute a hazard in use. Ethyl cellosolve and diacetone alcohol are less efficient but have the virtue of having very low vapor pressures, comparable to that of aniline. The vapor pressures of ethyl cellosolve, diaoetone alcohol, and aniline -. at 130° F are, respectively, 0.039 (2), 0.018 (2), and 6.070 pound per square inch. Isobutanol, normal butanol, toluene, nitrobenzene, and ethylene glycol are not considered efficient enough in their freezing point depressing capacities to merit further consideration. All the compounds mentioned above are not spontaneous with red fuming nitric acid under the conditions of use in jet-propulsion app~ratus, and hence have a retarding effect on the spontaneity of the reaction of aniline with the acid. Orthotoluidine reacts spontaneously, has a (stable) freezing point of 3.92° F (3), and forms two eutectics with aniline at 0.20 and 0.82 weight fraction orthotoluidine that exhibit freezing points of about o.46° and -25° F, respectively. In addition the vapor pressure of orthotoluidine is even lower than that of aniline, and it is readily available. Other compeunds have been considered. Furfu.ryl alcohol has a low vapor pressure, about 0.156 pound per square inch at 140° F, is available, is spontaneous in reaction, but pelyrnerizes when heated to about 338° F (4). This temperature m8¥ be lowered to 176° F if 0.1 per cent eydrochloric acid is present. Metatoluidine, which is spontaneous with red fuming acid, lowers the freezing point of aniline to -75° F if present in concentration of 0.426 mole fraction, according to Linard (5), but it is not available in large quantities. Orthotoluidine and acetone were selected for the studies presented in the thesis for the reasons here indicated,. Appendix A Bibliography 1. Timmermans, Bull. Soc. Chim. Belg., 37, 409 (1928). 2. Gardner, Indo Eng. Chem., 32, 230 (1940). !3. Bureau Int. des Et .. , Timmermans and Hennaut-Rola.nd, Jour. Chem. Phys. 9 32, . 604-7 ( 1935) • 4. 5. Ellis, Chemistry of Synthetic Resins, Vol. I, P• 522, Reinhold Pub. Corp, New York (1935). Linard, Bull. Soc. Chimo Belg., 34, 363 (1925). gg_, 1548. See also Chemo Abs., - J ,..1 r r - • I ,--I B C - --, - ---- A. .r.: - . D 1 J Schematic Drawin& of A.pparatu1 Uaed iJ1 Detel'Ulination of Free11114 Point. TABLE A-I. FIDilllZING POINTS OF MIXTURES OF .ANILINE .AND A NU1!J3ER OF ADDITIVE AGEHTS Added Material Weight Fraction Added Material Freezing Point (°F) o-Toluidine (c.p.) 0.0000 0.1021 0.1520 0.2002 0.2538 0.2998 o.4000 0.4999 0.6003 0.6999 0.8000 0.8999 1.0000 21.02 11.05 6.21 0.46 10.27 12.16 12.78 10.27 4.35 -6.36 -22.36 -18.52 -9.48 o-Toluidine ( 11 practical 11 ) 1.0000 -13.26 Acetom (c.p.) (1) 0.078 0.185 0.292 o.425 -1.3 -29.2 -103.0 Toluene (c.p.) 0.3000 -4.05 Turpentine 0.3001 * n-:Butanol (c.p.) 0.3001 -3.67 iso-Butanol (c.p.) 0.3000 -3.35 Nitrobenzene (c.p.) 0.3000 -4.• 32 Ethylene glycol ( c.:p.) 0.1163 0.2219 0.3014 +15.14 18.62 18.83 Diacetone Alcohol (c.p.) 0.0999 0.2000 0.3000 10.80 0.51 -15.54 • J!~orms two liquid phases below 54.2° F. -62.5 TABLE A-I. FREEZING POINTS OF MIX'l'URE OF AlHLINE .AND A NUMBER 0:B' .ADDITIVE AGENTS (cont 1 c.) Weight Fraction Added Material Freezing Point (OF) Ethyl Cellosolve (technical) 0.0999 0.2005 0.3004 +10.27 -1.74 -15.81 Methyl ethyl ketone (technical) 0.0998 0.2002 +5.41 -12.56 -36.03 0.0985 0.2100 o. z,025 +0.88 -19.05 Added Material Methanol (c.p.) o.3osa -33.3 40,------.-----..---------------- DIACETONE ALCOHOL LL -20 ETHYL CELLOSOLVE 0 ~ f-- -z 0 Q. -40 METHANOL 0 z N w w -60 er LL ACETONE -80 -IOO----f----------i~-- - f - - -- - + - - -----1 O. I 0. 2 WEIGHT FR.ACTION Figure A-I. 0. 3 0.4 0. S ADDITIVE AGENT Freezint; Points of Mixtures of Aniline and a Number of Additive Agents. 10 u.. 0 .... t- z 0 0. 0 l'.) z N w w 0: u.. -10 - 3 o~---......1-_ _ _ __.___ _ ___.__ _ _ _....___ _ ___ 0.2 0 .4 1.0 0.6 0.8 WEIGHT FRACTION ORTHOTOLUIDINE Figure A-II. The Freezing Points of Aniline-Orthotoluidine System. "" "" 0 0 u.. t-= -z 0 Q_ "" -10 (? z N KETONE w w 0:: u.. -20 -0- I SO-BUTA NOL -30 e n- BUTANOL © TOLUENE ? NITROBENZENE -40 ______..__ _ _ ___.__ _ _ ___.__ _ _ __ 0. I 0. 2 0 .3 0.4 WEIGHT FRACTION ADDITIVE AGENT :l!.,igure A- III. The Freezing Points of Aniline with Various Compounds. Appendix B THE MEASUREMENT OF BUBBLE-POINT PRESSURE , SPECIFIC WEIGHT, AND VISCOSITY OF THE REACTANTS Purgos_!!: To facilitate the design of units for the exchange of heat between the cold reactants and the hot jet u.nit, and to assist in the design of pumping equipment for these liquids, a study of the bubble-point pressures, specific weights, and viscosities was undertaken. Aoparatus and Method: Essentially the apparatus used was a glass capillary tube 0.080 inch in inside diameter, surrounded by an ebullioscopic bath. The pressure of the bath was regu.lated at any desired subatmospheric pressure by means of a manostat. The pressure within the capillary tube was measured by a flu.id pressure balance which had been calibrated against the vapor pressure of carbon dioxide at the ice point. The uncertainty of the actual pressure existing within the apparatus was probably less than 0.2 pound per square inch. However, because of the presence of uncondensable gases in the sample, which made the exact bubble po i ht uncertain, the error in the final results may be as much as 10 pounds per square inch. The copper-constantan thennocouple employed was: calibrated against a strain-free platinum resistance thermometer which had been calibrated at the Bureau of Standards, and its potential was measured by a White double potentiometer. The actual temperature measured was probably exact to within 0.05° F, but because of variation of the manostatcontrolled pressure on the bath and some superheating of the liquid in the bath, the final temperature for one point may be uncertain by as much as 0.3° F. A small hollow glass cylinder 0.061 inch in diameter and Oo5 inch long, containing an iron needle, was supported within the capillary by means of the field of an axially mounted electromagnet and was used to stir the sample durlng bubble-point pressure measurements and to ascertain the viscosity of the liquid sample. Figure B-1 shows a photograph of the assembled apparatus, and Figure B-2 indicates it schematically. The :fluid pressure balance was connected to the capillary tube through the oil-mercury-air interfaces in the vessel!• A mercury manometer was attached to ascertain the effective head of the mercury and oil in the apparatus with respect to the pressure balance, and a vacuum system was provided to pennit evac:.1ation of the capillary prior to sample introduction. The glass capillary was sealed to the metal part of the system by an unsupported-area packing gland l, and was contained in part in the vacuumjacketed ebullioscopic bath v.hich is shown within the electromagnet:!!.• boiler Q. supplied saturated gas to the upper part of the bath. The This fluid was condensed on the inside of the water jacket! and returned to the boiler by the external return!, thus preventing temperature gradients within the tube due to inhomogeneity of the fluid within the batho A small mercury-sealed stopcock provided at the bottom of the column, cooled by the water jacket l2.,was used as an inlet for a very small stream of air to prevent superheating of the liquid in the boiler, and also was used to change the fluid in the boiler. The ther- mocouple was introduced with a well extending down into the bath from the fitting Q, and the manostat was connected at g. Viscosity was measured by ascertaining the limiting rate of fall for the cylinder by measurements of time for different distances of falls while the fluid was held at the desired pressure and temperature. Two axially mounted electromagnets were used, the top one to hold the cylinder in the initial position and the bottom one to catch it below the final position. Both magnetic fields were tu.med off at the beginning of the fall, by a switch which also started an electric timer. When the cylinder had fallen a measured distance, the field!! were t1.1rned on and the clock stopped simultaneou.sly, while the cylinder was caught in the bottom field, permitting its return to the original position. The fall time was reproducible to within approximately 5 per cent. At least 60 fall times were measured for each state of the liquid. The instrument was calibrated by the known values for pentane (1) and water (2)9 It has been found necessary to ascertain the specific weight of the liquid and cylinder in order to obtain the viscosity from this type of instrument (3,4,5). This was done by the usual closed-pycnometer method (6), using a glass~enclosed magnetic E3€itator to assure equilibrium within the sample. The equation used to calculate viscosity was where 1 is the absolute viscosity of the liquid 9 f the specific weights of the cylinder and fluid, and f are empirically detennined constants. and f-s- are, respectively, t9 is the time of fall and c<. 9 This relation is derived from the usual equation of force that defines viscosity but with the incorporation of a ccJ.rrection for the inertial force exerted on the falling body by the flow of the liquid past it, which is taken as proportional to the density of the fluid times the velocity of the falling cylinder squarede It is estimated that the uncertainty of the viscosities meas.ired, from their absolute value 9 is about 10 per cent 9 about 0.3° F. The temperatures, as described above, are uncertain by B-4 Results: The red and white fuming nitric acids used were of compositions shown in Table B-Ib. The oxides of nitrogen present in the samples reported as nitro~en dioxide were determined by oxidation with eerie sulfate, followed by titration of the excess oxidizing agent with ferrous sulfates nitrogen present was ascertained by a nitrometer. The total The compositions shown in Table B-Ib were calculated on the assumption that only water, nitric acid, and nitrogen dioxide are present in the sample. Correlations of composition with specific weight are available in the literature, assuming, respectively, for red fuming and for white fnming acid that no water is present (7) or that no oxides of nitrogen are present (8). in the tableo These correlations are also listed In the case of the red fuming acid the analysis is believed more accurate, but in the case of the white fuming acid the composition determined from the specific weight is probably bettero The bubble-point pressures were determined for the two samples of acid described at a series of temperatures between 100° F and 300° F. B-I and Table B-Ia show the results. Figure In the case of the red fuming acid, decompe.s itien occurred slowly above about 200° F. For this reason a new sample was used for the values above 220° F, and the measurements were made as rapidly as possible. At the low temperatures tre bubble point was reversibly reproducible to within about 1 pound per square incho The white fuming acid did not show appreciable decomposition for temperatures up to 300° F for the times encountered in making these measurements. The specific weight at the bubble point is shown for various tempere.iures in Figure B-II and Table B-Ia. Corrections were made for the tem- perature coefficient expansion of the pycnometer, for the buoyancy of air on the weight of the sample, and for the small anount of acid in the gas phase. B-5 The pycnometer containing the sample was cooled to - 60° F and evact1ated to remove non-ccr.densable gases. No correction was made for the change of volume of the pycnometer due to pressureo It is believed that these measure- ments are good to 0.5 per cent. Viscosity at the bubble-point pressure, over the range of temperature 100° to 300° F, was ascertained. Figure B-III. The results are recorded in Table B-Ia and As mentioned previously, the change in specific weight of the liquid is accounted for in these measurements. The effect of pressure and temperature upon the dimensions of the falling cylinder and the enclosing cap illary tubing is not correeted foro The absolute accuracy is estimated to be 10 per cent. The vapor pressure (9,10,11,12,13,J.4) and Viscosity (15) of aniline are well known. The determination of these properties for one aniline-acetone mixture was made because of interest in acetone as a freezing point depressant in aniline& Table B-II contains the v&ues of bubble-point pressure for an aniline-acetone mixture containing 0.2975 weight fraction acetone. The measured values are from 3 to 25 per cent lower than calcttl. ated from Raoul t I s law. The values shown for 0.10 and 0.20 weight fraction acetone were computed assuming that the deviation from Raoult 8 s law is proportional to the concentration of acetonea The measurements of viscosity of this mixture were carried out. results are shown in Table B-III and Figure B-IVo The Appendix B Bibliography 1. Annual Tables of Physical Constants 9 4 9 sec. 381 (c), Frick Chemical Lab •• Princeton (1941). Dorsey, Properties of Ordinary Water-Substance, p. 183, Reinhold Pub. Corp., New York (1946). 3. Bridgeman, Proc. Am~ Acad. Arts Sci. 9 61 9 57 (1926). 4. Flowers, Proc. Am. Soc. Testing Materials, 14, 565 (1914) 0 5o Sage, Ind. Eng. Ohem., Ana.lo Ed. 9 ~. 261 (1933). 6. Dana. Jenkins, Burdick:, and Timm, Refrig. Eng., g (12) 9 387 (1926). 7. Klemenc and Ru.pp, Zeit. fur Anorg. Chem., ill, 55. International Critical Tables, e, 59. Lastovtsev, Rhim Maschinostroenie, .§. (3), 19 (1937). lOo International Critical Tables, .2, 221. 11. Landolt and Bornstein, Tabellan, Su.pp., Vol. 2 9 1309 (1931). Garrick, Trans. Faraday Soc., 23, 562 (1927). 12. Landolt and Bornstein, Tabellan, Supp., Vol. 3 9 2462 (1936). 13. Martin and Collie, Jouro Chem. Soc. (London), Po 2662 (1932). 14. Gilliland and Sherwood, Ind. Engo Chem. 9 26, 519 (1934) • 15., Steiner, Ind. Engo Chem. 9 Anal. Ed., 10,, 583 (1938). See also Figure B-1. Bubble-point Pressure and Viscosity Equipment . reservoir ·, Mercury Figure :B-2 Mercury manC11Jeter A Air ,...._ / <'apillary Glau ~ 1 ..--Acid Air ., sample -steel introduction . IIJ'IR9Dl I Plug tor sample To con1pressed- air oylinder To vacuum Schematic Drawing of Bubble-point Pressure and Viscosity Apparatus for Corrosive Liquids. Electrical contact circuit 011 To fluid pressure balance Dial gage - , D :.. -~~ ·---- •4 ~\::~i~:·t G Magnet C J~-· · I B E H To llianostat TABLE :B-Ia. VOLUMETRIC PHASE BERA.VI OR AND VISCOS:IT:Y OF FUMING NITRIC AC IDS :Bubble Point Total Pressure (lb/ in.2) Tem~r§iture (OF) White Fuming Red Fumi~ Acid Acid 50 204.4 155, 8 100 231.5 189.,1 200 265.6 228.2 300 290.0 255.2 400 309.9 275.7 500 326.6 292.8 308.5 600 Density Temperature (OF) Density (lb/ft.3) Whi te Fuming Red Fuming Acid Acid 75 92.20 98.90 100 90.57 97.,23 150 87.,43 93.45 200 84.,25 89.20 250 81.03 84.40 300 77.80 T.ABLE B-Ia.. VOLUW iETRIC PH.ASE :BEHAVIOR .AND VISCOSITY OF FUMING NITRIC ACIDS (cont 1 d) Viscosity Temperc1.ture (OF) Viscositi (centi~oises} Red Fu.ming Wm te Fuming Acid Acid Viscosit;t: (lb.secLft 2 ~ Vfn i te Fumirig Red Fwning Acid Acid 80 1.168 1.525 2.44 X 10-5 3.J.87 X 10 -5 100 0.990 1.240 2.07 X 10-5 2.592 X 10-5 120 0.842 1.010 1.76 X 10-B 2.111 X 10- 140 0.718 o.835 1.50 X 10-.... 160 0.617 0.690 1.29 X 10 1.442 X 10 180 0.536 0.575 1.12 X 10 -5 1.202 X 10-5 200 o.471 0.485 0.985 X 10-5 1.014 X 10-5 o.416 o.411 _i:; 220 0.870 X 10 .... 0.859 X 10 -5 240 0.371 0.352 0.776 lK 10-5 0.736 X 10-5 260 0.334 0.308 0.697 X 10-5 0.644 X 10 -5 280 0.302 0.269 -5 0.632 :x 10 0.562 X 10-5 300 0.277 o.579 X 10 .., 320 0.255 5 o.532 X 10- i:; -5 _i:; 5 1.745 X 10-5 • -5 T.AJ3LFJ l3-.lb •. COMPOSITIONS OF FUMING lITTRIC ACIDS !,hi te Fuming Acid Red Fuming Acid Density (gm/ cc) 1.4704 at 23.4° c. 1.5767 a t 25° c. Composition (wt.fra ction) Analysis N0 0.0054 2 HN0 3 ¾Ob a Reference (8). By Densitya ]y Analysis Density 0.2401 0.1818 0.8182 0.8634 0.8718 0.7259 0.1312 0.1282 0.0340 Density corrected for N0 , reference (7). 2 b By difference. c Assuming only HN0 3 and N0 present, reference (7). 2 C TABLE B-II o . BUilBLE POINT PRESSURE OF ANILINE-ACETONE MIX TUBES Va;eor Pressure, (lbL in.2) 0.10 Wt.Frac. 0.20 Wt.Frac. Acetonea Acetonea Temperature (OF) Aniline 80 0.005 0.373 100 0.016 0.622 l.019 1.432 120 o.044 1.006 1.694 3.289 140 0:0107 1.567 2.690 5.728 160 0.215 2.356 4.094 8.750 180 0.390 3.392 5.918 12.79 200 0.650 4.779 8.341 18.07 220 1.045 6.589 11.45 24.72 240 l.615 8.908 15.31 32.58 260 2.475 11.so 19.95 42.17 280 3.650 15.46 25.74 53.33 300 5.370 20.10 32.75 67.92 320 7.464 25.47 40.72 81.47 340 10.26 31.97 50.14 99.08 360 13.87 40.04 61.72 119.7 380 18.58 49.75 75.ll 142.6 400 24.38 60.78 90.10 168.7 420 31.62 73.81 107.2 199.1 440 40.55 88.80 126.6 233.9 b a Computed values. b From references cited in text. 0.2975 Wt.Frac. Acetone 0.2972 TABLE B-III. VISCOSITY OF ANILINE ACETONE MIXTURES CONTAINING 0.2975 WEIG-ET J!'RACTION ACETONE Ternperature (OF) Viscosity (cen ti poises) 80 1.32 2.75 X 10-5 120 0.761 1.59 X 10-5 160 0.536 1.12 X 10-5 200 o.423 0.885 X 10-5 240 0.354 0.740 X 10-5 280 0.308 o.646 X 10-5 320 0.278 0.581 X 10-5 360 0.253 0.532 X 10..., 400 0.238 0.498 X 10.., 440 0.224 0.469 X 10-5 Viscosity (lb. sec/ ft.2) _r:; _i:; 800 600 ·--· -I - - -- -- --····· ~ · - - - - · I I Vi I z -o-DECOMPOS ITION d (j) 400 0 er: w n. 300 aj _J RED FUMliG ACID ,... w er: 200 ::) (j) (j) ~ l SO - Q_ f- . z 0 100 I Q_ I w _J l! 80 (I) Cf) ::) m 60 I/ f 40 I V I V I I I/ / WHITE FUMING ACID 30 20 100 150 200 250 300 0 TEMP, Figure B-I. F Bubble-point Pressure of Fuming Nitric Acids. 350 (/) w u a. - LL l) ~ -w C) I ~- _J aj Q_ w 0: l) ::) u.. ~ L FUMING WHITE I I~ RED FUMING ACID 40 120 Figure B-II. 80 0 TEMP, F 200 240 280 320 Specific Weights of Ji\lrning Nitric Acids at the Bubble Point. 160 360 75 _ _ __.__________.____________..._______________ 80 84 68 92 961 100-.....,......---..-----~---------------...------------,------. Figure J3-III. Vi scosity o f ~,·ur.iing Hit:r.ic Aeids s t tho _;ubbl. e- point T'reasure 5 4 X IO t - - - - - - . - - - - - - , - - - - - - - . - - - - - - - - - . . - - - ----. 3 ~ ll.. . 2 0 (/) a: w Q_ u w -5 IX 10 (/) cO _J 8 .. >- I- 6 (/) 0 u l/) - > 0 4 0 WHITE -0- RED FUMING ACID FUMING ACID 3 2 X I-6 O '--_ _ _.....__ _ _...,___ _ _ 50 100 150 ~ _ _ ____,__ _ ____._ _ ____. 200 TEMP.°F 250 300 4.0 u) 3.0 I 0 X ~ 2.0 \ LL a (f) a: w 0.. u w ~ 1.0 (J) cri 0~ 0.8 ~ ....J >-'" 0.6 t- ~ (f) 0 u -.... ~ ---o-___ (f) > 0.4 ----- 0.2 80 160 Figure B-IV. 240 3200 TEM~ F 400 480 Viscosity of Aniline-Acetone Mixture Containing 0.2975 Weight Fraction Acetone at the Bubble Point. 560 Appendix C TEE IGNITION OF THE .ANILINE-ACETONE-AIR SYSTEM Purpose: One method of transferring the -reactants from their respective storage tanks to the jet unit is to maintain approximately constant the pressure of some gas in these containers and to control the flow of the reactants by valves and orifices in the reacta.'l'lt lines,. If air is used for the expulsion of these liquids from storage, there is a possible hazard from ·- explosion when acetone is used as a freezing point lowering substance in the aniline. It was desired to study this hazard by experiment and by calculationo Apnaratus: It is known (1) that the geometry of the apparatus employed affects the ignition of mixtures of acetone and air in at least two ways. The ex- perimentally determined limHs of composition for ignition may be made slightly wider by permitting upward propagation of the flame because of the gentle turbulence introduced .. Also, if the linear dimensions of the combustion chamber are all large, of the ord~r of o.5 foot or more, there are some borderline gas-phase compositions which will not ignite and. propagate throughout the volume but which may show ignition in smaller chambers .. It was desirable to build a small compact apparatus in which the temperature and compositions of gas mid liquid phases could be pred. sely controlled. Preliminfa ry experiments ~~th pure acetone indicated tbat a chamber 1 inch in diameter and 3 inches long gave ignition limits not appreciably different from those reported for geometry featuring upward propagation of flame and diameters of the order of Oo5 foot ( 1). Tbi s was borne out by comparison of the results for acetone in aniline with calculations. c-2 The apparatus employed is shown in Figure C-1. A steel chamber A is sealed at one end with a blow-out disc~ by fitting Q and at the other end - - by a. hollow fitting D which contains a spark plug with points shown at F and a connection for the introduction of air and the measurement of pressure at !• The pressure was measured by a Bourdon-tube pressure gauge, which had been calibrated against a fluid pressure balance standardized by the vapor pressure of ca.rbon dioxide at 320 F. Temperature was maintained constant by immersion of the pressure chamber in an agitated oil bath. Measurements of temperature were by a mercury-in-glass thermometer in the oil bath, which was checked at intervals by a standard thermocouple, which in turn md been calibrated by the strainfree platinum resistance thermometer from the National Bureau of Standards. Measurements of composition were checked 1:efore and after each series of measurements by ascertaining the index of refraction at 78.44° F on an Abbe refractometer. Procedure: Chemicals of the c.p. grade were used tbrQughout. A measured quantity of a controlled aniline-acetone mixture to be used was introduced into the chamber by removal of the fitting R of Figure C-1. The chamber we.s placed in the oil bath and allowed to reach equilibrium at a certain temperature. Air which had been dried and freed from ce,rbon dioxide by passage over calcium chlo ,ride, sodium hydroxide• and potassium hydroxide at a pressure above 4000 pounds per square inch was introduced into the chamber to the desired pressure. The ignition attempt was made by applying from 5,000 to 15,000 volts, depending on the pressure, to the spark plug from the second.ary of a 60-cycle transformer. responding energy dissipations were about 20 to 40 wa.tts. The cor- If no ignition C-3 was obtained, as indieated by a pulse on the Bourdon-tube gauge, the gaseous mixture in the chamber was changed by introducing air to a pressure of at least 100 pounds per square inch absolute and releasing for four timeso Then another charge of air was put into the chamber and 1 minute allowed for equilibrium to obtain before the next ignition attempto In general pressure was controlled so that the range covered went from non-inflammable to inflammable. Thl.s avoided introduction of more products of combustion than absolutely necessary. The 1-minute period for attainment of equilibrium was decided on after it had been ascertained that periods up to 103 times longer made no measurable difference. Results were distinctly not reproducible if the air-flushing process was omitted, even though ignition as indicated on the gauge had not occurred. Pressure changes of less than 2 per cent of the original absolute pressure were taken as due to heat generated by the spark and by reaction. the rate of which was below the range of practical interest. Each point was repeated using a fresh liquid sample to check the accuracy, and the composition of each liquid sample was checked before and after the point. The composition of the liquid phase for each ignition attempt was estimated by the change in refractive index and the number of air flushes and their pressures. Results: The experimental results are recorded in Table C-Io recorded are believed good to 0~1° F. The temperatures The weight fractions are probably not in error by more than 00002 since the change in any one liquid sample due to the introduction of air was not more than 0.0045 weight fraction acetoneo The pressures themselves are reliable to about 0.5 pound, lack of resolution of the limits probably being due to non-equilibrium conditionsa The incidental data used for the control measurements on the liquid sample are portrayed in Figure C-III. Since the vapor pressure of acetone is of about two orders of magnitude larger tr.an that of aniline, snd since the explosive limits of acetone-air mixtures are not very dependent on pressure and are known (1), it is possible to calculate the explosive limits of the aniline-acetone-air system if the fugacity of the acetone in the aniline is known at different concentrations and temperatureso These data are available at 68° F ,(::a) and cau Oe extended by calculations to different temperatures, using the heat of mixing given in reference (2), by the methods of Carlson and Colburn (3)o 'l'hese calculations assume that the effect of the dissolved air upon the fugacity of tne acetone in the liquid phase is negligible. The resuLts o:t" these calculations are shown in Table C-II, and comparison With the experiment values is made in Figure C-Ia Figure C-II is derived from Figure 0-I by tald.ng tne calculated or experimental limits that will give the wider rtange of inflammability. Appendix C Bibliograpcy lo 2. Jones, Harris, and Miller, Uo s. Bureau of Mines, Tech. Paper 544 (1933). Tables Annuelles Internationales de Constantes et Donnees Numeriques. VII, P• 251, Paris (1930). 3. Carlson and Colburn , Ind. Eng. Chem •• 34 9 588 (1942). .... J......----... . A-----;i-r • • n~e c..,1. Di~rarn of' apparatus TABLE C-I. IGNITION LIMITS FOR T.HE GAS PHASE OF HE'l'EROGEllEOUS MIXTURES OF ANILINE, .ACETOME, .AND .AIR U12~r Pressure Limit Total Wt.Frac. Temp. Pressure .Acetone (OF) (lb/in.2) in Liquid Lower Pressure Limit T<5tal""" Temp. Wt.Frac. (OF) Pressure Acetone (lb/ in. 2 ) in Liquid Ignition 130.54 0.3011 151.8 129.95 0.2993 35.0 No Ignition 130.45 0.3001 162.6 130.00 0.2987 30.3 Ignition 130.00 0.2003 91.3 130.07 0.1999 22.9 No Ignition 130.09 0.1994 100.9 130.05 0.1995 20.8 Ignition 129.86 0.1012 43.2 130.00 0.1013 14.3 No Ignition 129.86 0.1007 53.3 Ignition 80.10 0.3028 43.2 80.30 0.2996 14.3 No Ignition 80.22 o. :3022 48.3 Ignition 80.18 0.2006 22.9 80.04 0.1995 14.3 No Ignition 80.18 0.2011 28.0 80.00 0.1000 14.3 Ignition No Ignition 0.260 68 -- 24 0.0012 0,.0060 0.0407 -- 45 0.0023 0.0117 0 .0794 1.166 130 -- 22 o.oon 0.0057 0.0389 0.598 68 10 50 0.0026 0.0128 0.0874 1.284 100 0.2 19 95 0.0049 0.0245 0.1668 2.452 130 -- 37 0.0019 0.0093 0.0636 1.166 68 Limits of irflammability of acetone-air mixtures are reported as 0.0255 - 0.1280 mole fraction at 68° F. - Lower Pressure Limit of Inflammability (lb/in.2) 0.0005 Total Pressure 500 lb/ in.2 10 0 .0026 Total Pressure 100 lb/in..2 Upper Pressure Limit of Inflammability (lb/in.2) 0.0177 Total Pressure 14. 7 lb/ in.2 0.598 100 0.1 ANILINEl-ACETONE-AIR MIXTURES CAliHJLATED COMPOS I TI ON OF GAS PHASE AND IGNITION LIMI'11S OF Mole Fraction Acetone in Vapor Phase Fugacity of Acetone (lb/in.2) Temperature ( °F) Weight Fraction Aniline in Liquid Phase TABLE C-II. 16 80 0.0041 0.0205 001392 2.047 100 0.3 30 150 0.0077 o.0385 Ol2618 3.849 130 o--- -o- NON-INFLAMMABLE MIXTURE-EXPERIMENTA L 0 INFLAMMABLE MIXTURE~ EXPERIMENTAL UPPER PRESSURE LIMIT OF INFLAMMABILITY. CALCULATED LOWER PRESSURE LIMIT OF INFLAMMABILITYCALCULATED 2001-- - ---+-- - -- + - - -- - t--- - - - ; - -- --1- - - ---1 -<r 0 z 0 (/) a: 0~ w JOO,_----+-- - - - + - - - -- ,o,,__--+-_c~---+----~ v ~ a) ~ 0 80,__ _ _ _ _ _ _ _ __ ,... ~ ~ <vR_ 60----------+-f'------------------+ (/) w so~------ a: ~ ~ ~' ~«:-'?'G 40 -----'------+------+--,,,..._---+----- i 0 r .0 r-:!-· 0 /~ 30 / / ~v----++----+-------+--/-P- --+---- 01:J / / / 0 20----~--,,-----+--------/--+-----0- /-,-L---i- - - - 1 o· // / / / / / ri / 0·1/ 10"----.....L.-J.------'l'------'----~----..._--- 140 40 60 80 100 0 120 TEM~ Figure C-I. F Experimental arid Calculated Pressure Limits of Inflammability for .Aniline-Acetone Mixtures in Air at Var ious Temperatures. 300--- UPPER PRESSURE LIMIT OF INFLAMMABILITY LOWER PRESSURE LIMIT OF INFLAMMABILITY 2001-----+-- --+--- - - + - - - - - - -- z 0~ a (J) ,o er: w Q_ ~ 100 ~~ ill _J "'-0 80 Gy ~ w er: :) . (J) (J) ?" 0~ 60 /....."- 'v'G ~Q- w er Q_ _J 0"" 40 ,0 ~~ ~ f-- 0 f-- V / ~ 3-() o· / 1,, o· / 20 / / / / / / o~/ / 60 80 / / / 10 / / / / o') / 40 / 100 120 140 TEMP,°F Figure C-II. Critic ally Chosen Pressure Limits of Infl ammability for Aniline-Acetone Mixtures in Air a t Various Temperatures. . 1.58 ~ 1.57 1.56 \ ~ X w Q 1.55 ~ z w -> J- · 1.54 ~ u <( a:: LL. w a: 1.53 ~\ 1.52 1.51 1 1.50 0.05 0.10 0.15 WEIGHT FRACTION J?:'i. gu,rc-; ,.)-III. 0.20 0.25 ACETONE 1'he 1:n:5.e;c of Hefra.ction of An.Uine-Aceton0 \,{$. J: tU. t 0.S 1:1.t 70 .5° }' o  Figure 16. Continuous Reaction Apparatus{ FEACTANT FEED Pitlea TING COs kL CooLING WATER enn, 4, \wadgeaad Ss Zhe <= CL al y, “i, Brow -Our i Disa PRESSURE Gage th Figure 16. Contimuous Reaction Loeipment FRODUCT- SACIPL E Wa La belay we SPRING ~ LOADED - RELIEF VALVE \S A As | Av THERMIOCOU PL. -/ ZN Y CGBSERVATION © : — WINDOW & . . : - ; CAL/FE INST OF, TECH DEPT OF CHEM. ENG REVISION OF GALCIT CHAMBER 432 : | ee Fide Ge J [-3l- 42. ale | & OSA