Nuclear Rocket Studies

Page 2

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency Portions of this document may be illegible in electronic image products. Images are produced from the best available

Page 3

This document contains proprietary information and is submitted with the understanding that it will be held in confidence. All rights are reserved by Martin,

Page 7

II. Power Plant System Intergration................................................... 3 III. Supporting Studies.................................................................................. 31 C. Structural Design Considerations........................................ 49 D. Results of Reactor System Studies.................................... 51

Page 8

This report presents the results of nuclear rocket power plant studies conducted by The Nuclear Division of The Martin Company under Air Research and Development Command SR 150.

Page 9

The studies reported herein were undertaken to assist in determin ing the best way to flight test the first nuclear rocket. Since the nuclear system is yet to be designed, some flexibility is possible in the study work. Of the many test vehicle configurations possible, it was decided to investigate a two-stage device wherein the upper stage is nuclear and lower stage is a current ICBM booster. In this configuration, it is required that the nuclear stage possess the performance capability of propelling itself into earth-satellite orbit after first stage-burnout The study was directed toward determining the basic feasibility of developing such a nuclear stage. The system component which has greatest effect on basic feasibility is the reactor. Study results indicate that it is possible to develop a graphite-fuel-element reactor which meets the requirements of the test vehicle stage. The results indicate also that some rather remark able advantages accrue from the introduction of a moderate percentage of beryllium oxide into the core of such a reactor. Of great interest is the fact that it appears possible, with such core addition, to develop reactors with great growth potential. Results indicate that a reactor system can be developed in this fashion which can be flight tested atop an ICBM booster and which has the potential of powering a much larger rocket (in the neighborhood of one-half-million pounds thrust). This indicates a strong possibility of flight testing a nuclear Saturn second stage atop an ICBM booster and, perhaps, a nuclear Nova second stage atop a Saturn booster.

Page 10

The flight test portion of nuclear rocket development presents a challenging area of study. Interest in this area led to a series of studies by Martin Nuclear. The results of some of these studies are presented in this report. There are many possible vehicle configurations in which a nuclear rocket can be flight tested. Several of the more interesting are: (1) A single-stage nuclear test vehicle wherein the nuclear rocket is fired from the ground and propels itself into (2) A two-stage test vehicle with chemical first stage and nuclear second stage where the nuclear stage is started remotely at first stage burnout and propels itself into (3) A three stage test vehicle with chemical lower stages and a nuclear upper stage where the nuclear stage would be started after it had attained earth-satellite orbit. These possible approaches were examined in a preliminary fashion with due consideration given to such factors as: potential nuclear rocket missions; the current chemical booster program; and what was known of the Rover program. Results of this examina tion indicated the following: (1) The first available flight reactor will be basically a graphite system under development in the Rover program and will be in a relatively low thrust class--hence flight test planning should be concentrated upon this type (2) The first useful mission will most likely be an upper stage in the Saturn or Nova systems. (3) The single-stage test vehicle approach appears to offer no basic vehicle or powerplant feasibility problems. It does however, and, in addition, does not appear to be compatible with the power level of the earliest flight reactor planned in the Rover program. (4) The three-stage test vehicle approach appears to offer no

Page 11

basic vehicle or powerplant feasibility problems when the first two stages of Saturn are used. Any smaller chemicalboost systems, currently planned will not do the job. (5) The two-stage test vehicle is only of interest if it can use Saturn (in terms of current planning) and the use of Saturn provides the capability of more desirable orbital startup of the nuclear system. (6) The basic feasibility of a two~stage test vehicle had not been clearly established; nuclear rocket weight, primarily that of the reactor, is critical and a reactor design had yet to be evolved in the Rover program. The basic objectives of Martin Nuclear in conducting these studies is twofold: to make meaningful contributions to the national nuclear rocket program; and to prepare ourselves for eventual participation in that program. It appears from preliminary analysis that both of these objectives could best be fulfilled by directing study toward the determination of the feasibility of the two-stage test vehicle with an ICBM-type booster. Such study would provide the answer to a question of basic importance to the national program and would require a fairly comprehensive design study of the nuclear power plant, necessary for true understanding of such a device. The study was accomplished in two steps. The first was a vehicle study which yielded preliminary requirements for the nuclear stage. It has been reported previously (Ref. 15). It indicated that the nuclear power plant would have to weigh less than 10, 000 lb and produce 67,000 lb thrust if it were to attain orbitting conditions from atop a Titan first stage. The study was somewhat conservative in that 3140° F average reactor exhaust hydrogen temperature was The study reported herein is concerned with the power plant, with emphasis on the reactor which is by far the heaviest component in the system. As in most studies of this nature, the work, as it pro gressed, uncovered many associated problems of interest. Some of these in the areas of reactor startup and radiation heating are of sufficient magnitude to effect system feasibility. The lack of treat ment in this report is not intended to indicate lack of importance. It is planned to give these problems the attention they deserve in the future and report on them at a later date.

Page 12

II. POWER PLANT SYSTEM INTEGRATION This chapter presents the results of power plant parametric studies wherein the weight and performance of all system components are evaluated as a function of several common variables, and overall power plant systems are evolved, also as a function of these variables. From vehicle studies conducted previously (Ref. 15), approximate thrust requirements were obtained. From this and an assumed value for average reactor exhaust gas temperature of 3600° R, a flow rate of 103 lb per second and a reactor power level of 1300 mw were obtained. These values were held constant throughout the power plant studies. In an iteration of vehicle performance using power plant data generated in this study, however, a variation in propellant flow was examined to determine its effect on the trajectory of the vehicle. From the preliminary analysis described very briefly in the Introduction, it appears that the earliest flight reactor may not fit the earliest mission from a power-level standpoint, unless the nuclear systems are ganged or some growth potential is provided in the reactor design. A rather obvious way to provide reactor growth potential in power is by variation of propellant pressure and flow rate. In this fashion, it may be possible to flight test an over designed reactor at very low pressure and use it eventually in a higher powered mission role at high propellant pressure. With this thought in mind, it was decided to look initially at low pressure systems. This, in turn, prompted study of tank-pressurized feed systems which would appear to offer the advantage of simplicity for flight testing purposes in the low pressure regime. It was felt initially that failure to obtain useful systems in the low pressure, tank-pressurized regime would result in a shift in em phasis to pump-pressurized systems. The study, however, yielded useful tank-pressurized systems and there was no need to conduct pumping system studies to accomplish the current objectives. In the study, uncoated graphite fuel elements were assumed and the hydrogen propellant was assumed to contain two mol percent of methane. Recent information indicates that this scheme is currently in disfavor at Los Alamos and that coated fuel elements and no methane are anticipated for the early flight system. The effect of this assumption has not been evaluated, but it appears to be some-

Page 13

thing of a standoff with the requirement for larger tankage offset to some degree by the greater specific impulse of the all-hydrogen In establishing the physical and performance characteristics of the exhaust nozzle for application to the optimization study, consider ation was given to the use of both the plug-type nozzle and the con ventional convergent-divergent nozzle. When designed to the same conditions, the two nozzle types provide virtually the same specific impulse in vacuum operation and the advantage of one over the other lies primarily in their relative weight relationship. At first inspec tion, the plug nozzle appeared to have merit in that the plug was to serve, at least in part, as a housing for the reactor. This arrange ment would serve to make a more compact propulsion system which would minimize the weight of the second stage supporting skirt. In the final analysis, the plug nozzle and its associated structure proved in every case to be heavier than the conventional convergentdivergent nozzle. On the basis of this finding, the plug-type nozzle was eliminated from further consideration. To provide minimum weight, the conventional convergentdivergent nozzle was contoured according to the method of charac teristics for shorter length and parallel exit flow to provide a thrust equivalent to that of a conical nozzle having a divergent half-angle of 15° (Refs. 2, 13 and 14). The throat region of the nozzle was to be cooled by the propellant, this portion of the nozzle being of stainless steel tubular construction. Cooling was applied only to that portion of the nozzle where the wall temperature, without cooling, would exceed the allowable working temperature of the metal. The cooled portion encompassed approximately one-third of the overall nozzle length. The remaining portions of the nozzle were uncooled and were assumed to be constructed of 0.064 in. stainless steel sheet. This method of construction is such that, for a given exit diameter, the weight of the nozzle is essentially independent of the chamber or reactor exit pressure within the range of pressures considered. The weight of the nozzle varies significantly only with nozzle exit diameter as can be seen in Fig. II-l. The stagnation temperature was assumed to be 3600° R and constant throughout the study. It followed, therefore, that the nozzle thrust or specific impulse could be expressed as a function of nozzle area ratio for vacuum operation. The specific impulse-area ratio relationship, as used in the study, is presented in Fig. II-2. For the purposes of the system analysis, the nozzle weight curve (Fig. II-1)

Page 14

and the nozzle performance curve (Fig. II-2) were combined into a single curve (Fig, II-3) expressing nozzle weight as a function of specific impulse and nozzle total pressure. This latter curve com bines the weight and performance aspects of the nozzle and facilitated the determination of nozzle effect on the overall system. The propellant system for the nuclear powered second stage rocket is of the tank-pressurized type using liquid hydrogen contain ing 2 mol percent liquid methane as propellants. The propellant tank diameter was fixed at thirteen feet and hemispherical domes were used to enclose the tank ends. Titanium alloy A-110AT was selected as the material to be used for tank structure. The selection was based upon data available on the mechanical properties of high strength steels, aluminum, magnesium and titanium alloys at cryogenic temperatures. These data were taken from Ref. 1 and are presented in Figs. H-4, II-5 The titanium alloys exhibit the highest strength-to-weight ratios at -25 3° C and retain sufficient ductility to be acceptable as useful materials. Titanium alloy A-110AT appears to be the most promis ing material at these temperatures as it is superior in yield strengthto-weight ratio and retains sufficient ductility to permit 3% elongation at -253° C. It is also readily weldable and does not require reheat treatment to realize its high strength properties at low temperatures. The principal factors in the design of rocket propellant tanks are the axial loads due to internal pressure and power plant thrust, bending moment loads and the hoop tension stress due to internal pressure. It was assumed that, at the three critical times of t = 0, t at maximum dynamic pressure and t at burnout, the axial load was always a tension load and that columns or stiffeners were unnecessary. Hoop tension was then considered as being the factor determining the skin gage. For a modulus of elasticity of 17 x 10b and a yield tensile strength of 260,000 psi for the A-110AT alloy, skin gages were established for a range of pressures from 60 to 180 psia for an assumed factor of safety of 1.25. The material gages calculated for a tank pressure of 100 psia were established as the minimum for safe and reasonable handling, and were held constant at pressures less than 100 psia. Having thus established the dimensions of the tanks, the gage and material to be used, the governing design cri teria, and drawing upon the experience gained in past tank designs, the weights were estimated for various combinations of propellant

Page 15

quantity and tank pressure. Figure II-7 is a plot of this relationship. The abrupt change in the slopes of the curves at 100 psia reflects the limiting of the structural gages to the values required at 100 psia. It will be noticed that the tank weight increases with pressure at a rate greater than can be accounted for by hoop stress increases alone. The increase in weight is influenced to a great degree by increases in tank length which are dictated by an increasing propel lant volume (for a fixed propellant weight) for increasing tank pres sures. The capacity of the tank is such that the proper quantity of propellant is contained within the tank when the temperature reaches a point where boiling takes place at each pressure. The density of the liquid …