Iíve appeared in several public forums recently (the ISDC, Spacevidcast and The Space Show) talking about nuclear rockets and that prompted many questions and comments. Rather than answer them piecemeal, I thought it best to present an introduction or primer on key technical and economic aspects of nuclear rocketry not only to answer these questions but more, to allow the reader to get ďnuclear rockets in his or her bones,Ē to get a visceral or intuitive feeling of what a nuclear rocket space program can do. That really seems to be the main thrust behind all the comments. Other than manned Mars, what can it do? And, I will add, what can it do for me? The simple answer: create a new space program, but only with new thinking. In discussing that, I also say the reader must then take action to create that new space program, one open to the common man.
With this purpose then, I will only summarize points made in my two books and refer the reader back to either for more information. My first, To the End of the Solar System, is a political and technical history of the Rover/NERVA program from 1955-1973 while my second, The Nuclear Rocket, is a broad policy argument for how and why to restart the program: it holds nuclear rockets revolutionize the space program. They are the game-changer. The first is almost out of print, but Apogee Press still has a few copies; the latter is available from Apogee or other booksellers, such as Amazon.
This introduction has seven parts.
Iíll cover the basic stuff first. A chemical rocket burns a fuel and oxidizer and then exhausts the gas out a nozzle to go forward. Most nuclear rockets use uranium to heat a gas and then expel it out the nozzle to go forward. (Newtonís second law applies to both). So one burns a fuel and oxidizer, one heats a propellant gas, e.g., hydrogen. More precisely the isotope U-235 thatís been enriched to 93% is used and itís called highly enriched uranium (HEU) or weapons grade uranium. Here the U-235 atom is fissioned or split in two, creating vast amounts of heat or thermal energy, isotopes with molecular weights half that of uranium and radioactive particles such as gamma rays and neutrons. Three classes of nuclear rockets exist (excluding fusion, anti-matter and fission fragment-nuclear bomb types). First, in the solid core uranium is mixed with a matrix material, such as graphite, and fabricated into fuel elements. Second, in the liquid core uranium is allowed to melt and is contained in a pot-like structure, like molten iron in a crucible. Third, in a gas core uranium is turned into plasma and is contained in a fluorescent bulb-like structure. In all three hydrogen is used as the working fluid/propellant. Liquid and gas cores promise significantly higher specific impulses (Isp), well into the thousands, but at the moment only the solid core is feasible. It is the subject of my two books.
The Nuclear Rocket examines a flight profile NASA banned in 1960 and except for a few reports it hasnít been studied since. Thatís over half a century. Itís the flight from the earth to LEO and then return from LEO. Breaking this ban opens space up to the common man, but NASA banned it because the newly created agency simply feared a public relations headache and not for any solid technical reason. Nonetheless, this flight profile was common before the ban and started with LAMS-1870 ďThe Feasibility of Nuclear-Powered Long Range Ballistic Missiles,Ē issued in March 1955. Here the Los Alamos National Laboratory argued a ground launch of a nuclear rocket lacked a payload advantage, but boosting the stage to 50,000 feet, where it fired, had a large one. All Iíve done is go back to this premise and reexamine it from todayís perspective Ė so this is not original on my part. What is original is that Iíve taken the next steps to see the effects this launch profile could have on the space program. They are stunning, it transforms our current space program from a narrow, government run, elitist effort to a broad-based and democratic one that can change society and our international order by bringing peace to the planet. The latter thought seems particularly preposterous, but it goes back to some of the most brilliant scientists of the 20th century. Oh! You may sniff, more self-righteous scientists and their scientific predictions. Mostly wrong! But these guys, I hold, had a proven track record. They were spot-on many times. So again, I claim no originality with their insight. What is original is that I set forth in The Nuclear Rocket a straw man argument on how it might be achieved to provoke discussion and dialogue. I hope you read it, as I cannot mention it here.
Iíll say this from a different angle so the reader may better grasp my argument. A chemical rocket space program is premised on what I call ďmisson-itusĒ thinking. In other words, since going into space is costly with chemical rockets, a mission must be developed first and they can be divided into three broad categories: science, national security, or commerce. Most are government sponsored and follow a pattern: an agency develops and approves the mission, then it goes to the President for approval, and then to the Congress for approval and the authorization and appropriation of funds. Afterwards, contracts are issued for the private sector to provide the rockets and payloads. This takes time, years actually. The private sector follows a similar pattern though the time involved can be shortened since they do not follow the government approval and contracting process. Moreover, nearly all rocket launches are one and done and this is basically true for the space shuttle, which has to be rebuilt/recertified before another launch.
To this we must add the inflexible rocket equations or we can say the inflexible rocket equations cause all this. Its most critical part is: V is proportionate to ÷T/M. (Sorry, I donít have the proportionate symbol on my computer). Here V is the expulsion velocity of the exhaust gas and T is its temperature and M is its molecular weight. The best fuels are LH2 and LOX; they burn ~4000įC and have a molecular weight of 18. This gives a specific impulse (Isp) of ~450 seconds. All other chemical propellants have much poorer T and M properties, and so have much lower specific impulses, ~330 for solids and kerosene and LOX. A few exotic chemicals offer Isp slightly higher than 450, but they are quite nasty or toxic to handle.
So the bottom line of T/M is this: ~90% of a rocket sitting on the launch pad is fuel and oxidizer; nothing can be done about that because you need that much to climb out of earthís deep gravity well and you canít make them lighter (you can make some denser and thus save on tank weight). Their molecular weights are set and newer, lighter elements are not forthcoming on the periodic table to change this. Of the remaining ~10%, the rocketís structure (tanks, engine, nozzle) takes up ~8%. This leaves ~2% for the payload or the payload fraction. With so small a payload fraction, even ounces are precious and great effort is spent to use every ounce effectively. So payloads cost millions or billions.
(This dismal situation is quite similar to the Pony Express of 1860 where each pony had to weigh less than 900 pounds, each rider under 110, leaving 55-pounds for the saddle and bridle, riderís clothes and a 20-pound mail sack for a total of 1065-pounds. In rocketspeak, the Pony Express had a 1.8% payload fraction. So what we have today is a Pony Express space program; T/M prevent it from gaining larger payload fractions, itís impossible when ~90% of a rocket on the launch pad is fuel and oxidizer).
So we have ďmission-itusĒ thinking combined with static chemical rocket development limited by T/M; both combine to give a very stagnant space program. To break out, to put the ponies out to pasture, one must increase T while lowering M; you must focus on the 90% of the rocketís weight Ė the fuel and oxidizer - and not on the 10% as has been done for the last half century. That is precisely what nuclear rockets do Ė deal with T/M. And here, even with the humble solid core, the process becomes very dynamic, with major gains made from one-generation system to another and with many generations possible. Thus itís easy to see each generation system will have an increase in payload fraction - it wonít be stuck ~2% - and this promises to lower the cost to LEO to around $100 per pound. And as even more advanced solid cores are introduced, this will go even lower. Iíll soon show how.
The Rover/NERVA experience reinforces this point: the best thinking of 1960 had been radically overturned by a decade of R&D in which more and more ways to improve continually appeared. These were simply not on the horizon in 1960; in fact, they were not even dreamed of. Moreover, forty years have passed since the program ended, time in which advances have been made in many technologies that can be incorporated into todayís program. The bottom line is one can expect a very dynamic program with ever larger payload fractions. As a consequence your mindset must be dynamic, expecting improvements in quick order and not moribund as with chemical rockets.
(Let me say I quickly saw this dynamic situation, when applied to launches to and from LEO, meant the old 1960s model of the government letting contracts to private firms such as Westinghouse and Aerojet to develop a nuclear rocket wouldnít work. That would privilege a few while putting everybody else out of business; these chemical rocket firms simply couldnít compete. In other words, Iíll shout this at 120 decibels: it will not be a government-funded effort! Hereís one example why. In one of the few exceptions to the ban, in 1968 Stan Gunn at Rocketdyne studied replacing the J-2 third stage of the Saturn V with a first generation NERVA I (75k thrust, 825 Isp, 15,000-pounds) and saw it could almost double the payload to LEO, from up to 250,000 pounds for the all-chemical Saturn V to up to 500,000 pounds. However, because of its ban NASA never embraced this and it died quickly. You donít have to be a rocket scientist to see this enormous potential, even with a first generation system; that is why I developed the NucRocCorp concept, (itís a horrible name I know) a government-industry structure with industry in control and providing most of the money, to provide for participation by all rocket firms within the US and overseas, if they wish, and spent several chapters introducing it in The Nuclear Rocket. Please read it. Itís a straw man to provoke discussion and dialogue. Reading it will also disabuse those who think the government will fund the program. No! No! No! Our government is broke. Thus, we must think differently and develop a structure that has private funding, i.e., a NucRocCorp).
First, perhaps the hardest thing to do in nuclear rocketry, at least with the solid core, is keeping it inside the pressure vessel and not flying out the nozzle. This is because hydrogen will enter the top of the core at a very high pressure and leave in a very short distance (e.g., four-feet) at a very low pressure. This core pressure drop creates a force that tries to blow the core out the nozzle. To handle it, Los Alamos studied dozens of core concepts from 1955 onwards and I list some in To the End of the Solar System so no need to repeat that here. All had problems but in 1960 the B-4 core concept emerged and while debated fiercely, it soon won out. However, even it had problems and was not successfully tested until September 1964. Thatís almost ten years since the program began. So if you see proposals for new nuclear rocket concepts, featuring fuel elements with fine properties, such as carbides, take it with a grain of salt. Their cores may consume a lot of time and money and at the end of the day prove unworkable.
Thus in my analysis, I advocate the B-4 core: it will stay in the pressure vessel and more than that, it is extremely flexible and can be made very swiftly and easily into different sized engines and it has a lot of growth potential. In other words, I want to shout this at another 120 decibels: itís mature, just take it off the shelf and dust it off; little R&D is necessary to begin flying (the gimbals probably the biggest challenge)! This allows the R&D to focus on ever more advanced B-4 systems or on other concepts. With this being so, all talk of nuclear engines costing billions and billions to develop is nonsense if the B-4 is selected. What will cost are test facilities. I havenít been to Jackass Flats in the Nevada Test Site (where nuclear rockets were tested) for several decades, but others who have been there more recently gave a bleak assessment of the situation, saying it might be better to start over elsewhere. I donít know. With high unemployment in Las Vegas, refurbishing Jackass Flats and turning it into an active nuclear rocket R&D center might be highly desirable politically. I gave my estimate for refurbishing it in The Nuclear Rocket; itís a starting point from which to begin analysis and then debate and dialogue and once that happens other sites would also be thrown into the mix.
I donít disagree with these other concepts or carbide ones (which need a core design different from the B-4) and could see them being evaluated after B-4s begin flying, but I also feel they seem to be premised on ďmission-itusĒ thinking. Theyíre developed for specific missions such as manned Mars or lunar bases. When you break the ban, you must concentrate on building its infrastructure, one that allows the public to use it, and you must do this as quickly, else wise the program will lag and then critics will be merciless. These citizens will fund the missions and those will be much, much broader than science, national security and commerce. So a good one-sentence summary of the difference between nuclear and chemical rockets is that the former permit the common man access to space while chemical rockets limit it to government or corporate elites. Thatís what itís all about really, though I must say the government will still need launches for its purposes, particularly national security, but NASA will have a fundamentally new mission. Read about it in The Nuclear Rocket.
Focusing on infrastructure and not missions is like the process the Navy followed in building nuclear subs. Launched in 1954, the Nautilus was first, to demonstrate the technology, but it didnít have any real military utility, some torpedoes only. It was a learning tool; to think the Navy thought of Poseidon and Trident missile carrying subs when it started building the Nautilus in 1948 is absurd. Rather, the Navy knew it would have been ridiculous to follow the diesel/electric sub model that had to surface every few days to recharge the batteries. Instead, the Nautilus allowed the Navy to gain experience and from that to develop the policies, practices and procedures and then the physical structures and mix of workers to make it all happen Ė the infrastructure in other words Ė to use this new technology to its fullest. Many missions then followed. I suggest this focus is the model to follow once the ban is broken. Build your democratic infrastructure, the common man will determine its missions. Also read about it in The Nuclear Rocket.
Second, I say start with small engines, then after a decade, after experience is gained and an infrastructure somewhat developed and matured, go to larger ones. This is a complete reversal of the thinking of the early 1960s where the focus was on building gargantuan systems for manned Mars missions, the most obvious justification then for the program once the ban was established. They were stupendous, 10,000 MW (500,000 pounds of thrust, 12,000 MW (600,000 pounds of thrust), even 20,000 MW (1,000,000 million pounds of thrust) and utterly impracticable and they were proved to be a product of chemical rocket thinking applied to nuclear, of diesel/electric subs to nuke subs. Some of this thinking disappeared as the 1960s wore on but unfortunately manned Mars remained as the major justification for a nuclear rocket. This thinking just wonít die and has been a poison pill for half a century; it is easily attacked and killed politically if itís government funded. Itís happened repeatedly and within the bowels of the space bureaucracy manned Mars planning is probably happening now. Wonít those advocates ever learn?
Third, I propose a cargo plane launch. In the early 1960s, large heavy lift planes, such as the C5-A and Boeing 747, didnít exist and even when they did, their ability to launch a large rocket was unproven until the Air Force launched a Minuteman missile out of a C5-A in 1974. By that time Rover/NERVA had been canceled. Cargo plane launches have many advantages. They eliminate the need for costly fixed installations such as the Cape. If the private sector had to pay for this, its launch costs would be outrageous and probably prohibitive. With a cargo plane, all you need is an airport. They also allow maximum protection for the public, as the actual launch would be in an isolated ocean area, such as the vast Pacific Ocean, where there are few people and ships for thousands of miles. Moreover, security measures can be very effective, as the planeís actual destination for launch can be tightly controlled. The US has many islands in the Pacific that can be used as advanced support or staging bases for the launches as well as for the recoveries of the nuclear engines from LEO. They also can provide shelter in bad weather. With such a vast, uninhabited area, many launch corridors are possible so if conditions are adverse in one corridor, others may be selected to avoid canceling the launch and waiting until conditions are right. The meteorological conditions of the southern Pacific have been studied extensively, particularly during the atmospheric nuclear weapons testing time period (1945-1963). Many reports exist on the effects of radioactive particles in the atmosphere. This includes UN studies in 1958. However, the amount of radioactive effluents coming from a 10-minute flight to LEO, where the engine begins full power operation above 100,000-feet, and then flies to LEO, will be almost invisible when compared to those coming from a single one-megaton atmospheric test, or a smaller 10 kiloton test (a Hiroshima type bomb), let alone the hundreds conducted over this 18-year period.
Fourth, the C-5ís cargo capacity will be a determining factor in developing the nuclear stage, at least initially. Weight will not be a problem, it can carry ~300,000-pounds, but the cargo bayís length, height, and width dimensions are critical because LH2 tanks are large. Modifications can be made to a C-5, but as experience is gained, it may be desirable to design a cargo plane just to carry nuclear stages, either inside its cargo bay or underneath so the stage is dropped like a bomb. Moreover, the Air Force is working for NASA on extracting 90,000-pound Ares rocket stages out of a C-17, so extracting a nuclear stage of comparable weight does not appear to be a problem.
Five, the B-4 was built and tested in various core diameters, from 55-inches to the most common size of 35-inches to the 20-inch. All were 52-inches long. There was no unsolvable problem in going up or down in size despite the fact that each was pushing the state of the art and testing new ideas. In fact, after the B-4 core was proven sound in 1964, Los Alamos and Aerojet/Westinghouse tested nine between 1965 and 1969. Thatís one every half year. Even more could have been tested if the budget was not declining and if Test Cell-C was not out of commission for a year and a half to add a million gallons of LH2 capacity. Thus, I confront the handwringers who cling to the myth ďItís so difficult, costly and time-consuming to build nuclear rockets.Ē Utter nonsense Ė one every half year! Itís easier than chemical rocket engines, particularly with the computer modeling tools now available. Los Alamos and Aerojet/Westinghouse pioneered the use of modeling in their tests, taking the data from a run and feeding it into the model after every test. Over time, actual tests became more of ďdemonstrationsĒ of what was learned rather than ďexperimentsĒ to learn new data. In fact, ANSYS, a $3 billion market-cap corporation listed on NASDAQ, started from the NERVA program and can do just that.
I hold the 20-inch Pewee is the reactor to develop into a small engine. It took 19 months from when Washington approved it to its test and it was the most advanced of all the systems tested though its purpose was to test fuels and not be developed it into a flight rated system. It held 534 hexagonal fuel elements, each ś of an inch from flat to flat; 402 of these elements contained uranium while the other 132-zirconium hydride fused tubes to promote criticality. It had a design power of 500 MW or 25,000 pounds of thrust, ran on LH2 and weighed 6000 pounds. In an hour run, it reached and stayed at 2265įC three times for twenty minutes each, severely shocking the fuel. The fuel development goal by now had changed from up to a half hour in 1960 to one hour at full power with several recycles (stops and starts) in 1964 to 10-hours at full power with 60 recycles.
(In 1960 this was undreamed of and transformed NERVA into an 18-wheeler in LEO, going out, coming back, and going out again until 10/60 was exceeded. This thinking is still valid. I modify it in The Nuclear Rocket with the idea of creating a solar system transportation plan with different nuclear rockets operating like a bus, train or airliner, going out and coming back at regular intervals to various destinations in the solar system. This is not ďmission-itusĒ thinking where the elites define a mission first; here the schedule/destination would be published and the cost of the ride. The common man would fill up the payload for the trip out and back. I think to do this right would take teams at least a decade, but those adventurous souls who might want to take a peek at such a solar system transportation plan could start with a simple exercise. Assume nuclear rockets with specific impulses of 1000, 1200, 1400, 1600 and 1800 seconds and then calculate the speed at which they will travel to different destinations and then return to earth. The results will be awfully rough and probably wrong, but it will get your minds thinking in a different direction. But for those who Iíve shocked into disbelief at a solid core with 1600 or 1800 seconds, I say hold on, wait until you finish this paper. Thereís a lot of life in the humble solid core, it doesnít top out at 925 or 1000 seconds of Isp).
Pewee had a calculated Isp of 845. Four others were planned, and one was built and out at the test stand when the program was cancelled. The last three were to have scrubbers to remove radioactivity from the plume. I will explain them shortly. Of all reactors tested during Rover/NERVA, it had the best power density ever, a term I will also explain shortly to those who believe the bogus fat, heavy reactor or lousy T/W ratio arguments.
There is no technical reason why Pewee cannot be developed into a flight rated system quickly, easily and inexpensively Ė remember it took only 19 months - and it will progress quickly from a first to a second, third and fourth generation system. No real new R&D has to be done; in fact, Rocketdyne had to derate one of its turbopumps for Pewee; no problems exist for the nozzle also. Though it might be 1960s era technology, it has a lot of room for growth, as I will illustrate.
So now letís take stock and see where we are. I advocate breaking the taboo on using nuclear rockets to reach and return from LEO, but believe in starting with small engines first and focus them on building an infrastructure to use them. I also believe in a cargo plane launch in isolated ocean areas - the south Pacific - where public safety risks are nonexistent. Conceptually, the launch profile features three steps: the cargo plane taking the stage to ~50,000-feet, whereupon it is discharged from the plane. Two or more solid boosters then fire to take the stage up to the upper part of the stratosphere (about 6 to 32-miles up) or somewhere above 20-miles or 100,000-feet whereupon they separate and the nuclear engine then fires to LEO. This solid booster lift does four things:
Now Iíve stated nuclear rockets are a dynamic technology that changes not only the economics but also the fundamental nature of the space program, so your mindset must be equally dynamic. I will show how in three areas: specific impulse, power density, and other nuclear or non-nuclear improvements.
Specific Impulse (Isp). Increases in Isp mean increases in speed and thrusting power (I exclude electric/ion propulsion from the discussion and concentrate on reactors with hydrogen as the working fluid/propellant). For a short flight to LEO, increased speed is not desirable but increased payload is. Recall the Rocketdyne Saturn-Nuclear study is mentioned earlier. For missions starting in LEO, tradeoffs must be made, more speed for a mission or the same speed, but with a bigger payload. This is a real big deal. Some hold once nuclear rockets reach Isps of 1500, it opens up the inner planets to colonization; others say itís 3000 seconds. Some say if gas core nuclear rockets reach 8000 seconds, it means Mars in a month. Iíll say this differently and I hope you get this principle into your bones: as nuclear Isp increases the solar system shrinks time wise. And we are able to do more and more and more faster and faster and faster. We no longer are running a pony to Pluto with a 20-pound mail sack.
Since hydrogen is used, it is the lightest element and has a molecular weight of 2 and this means any increases in Isp must come from increases in T, temperature. (I will discuss an exception to this shortly). For simplicityís sake, letís just say 800 seconds of Isp means a T of 2000įC while 1000 seconds means a T of 3000įC. 4000įC appears impossible, as nearly all known materials melt by then, but this should not be viewed as absolute, because materials science does advance. During Rover/NERVA, fuel elements were made with graphite and it sublimates, or turns from a solid to a gas at 3700įC, and hydrogen corrodes it swiftly. Think of how quickly a sugar cube dissolves in a cup of coffee and youíll get a visceral idea of the problem. Thus, the stories some speak of today Ė reactors pouring radioactive plumes into the environment. ďOh, the horror of it!Ē To this I say, take a deep breath, hold it, and then exhale slowly and then read my appendix in To the End of the Solar System on Safety and the Environment. If it were that bad, the EPA would not have given kudos to NASA in 1974 when it conducted a wrap-up study of all of the programís effluents, holding they were under the guidelines. Something is vastly out of whack then; either those who believe the ďhorrorĒ stories are greatly misinformed or the EPA was knowingly fraudulent and dishonest. You decide.
When reactor testing started in 1959, little was known about protecting graphite from hydrogen and as a consequence, the fuels eroded swiftly and runs were around 10-minutes. So the sugar cube analogy is appropriate. And at this time, weíre talking about 800 seconds of Isp or operating at 2000įC but as the 1960s wore on, the materials people began producing better and better-clad fuels, able to last much longer in a hot hydrogen environment. Indeed, the NRX-A6 ran for 62 minutes at full power, going through a million gallons of LH2, and the fuel elements suffered little loss of uranium. In other words, the sugar cube remained almost intact. That was 1967 Ė only 8-years from 1959. After 1967, the goal became the 10/60 Iíve mentioned earlier, but at higher temperatures. So while the program was going on to solve harder and harder fuel objectives Ė they were now talking graphite fuels with T between 2600-2800įC - it was in its death throes. Moreover, just before it ended, Westinghouse exceeded the 10/60 objectives in furnace testing in 1972, going 64 cycles. What a stunning change from 1959!
After the Nixon Administration killed the program, Westinghouse and Los Alamos fuel development stopped, but Y-12 continued working on graphite for other purposes. A Y-12er who worked on graphites during that period said ďgraphite has a lot of life in it, donít count it out.Ē This means with a renewed program, temperatures of 3000įC+ are possible. This means an Isp of 1000+ seconds. Carbide fuels, however, promise even higher Isp, and the program was heading in that direction when it was cancelled. Meanwhile, the Russians reportedly ran carbides well over 3000įC, high enough to give an Isp ~1100. However, thatís an unconfirmed Russian news article from 1995 so it must be viewed with reserve. What theyíre doing now I donít know.
Power density. This term in not found in chemical rockets and can be defined as the amount of power derived from a given core volume. To help understand this, I find the best analogy is from car engines and Iíll use my good friend Stan Gunnís 1960s something Corvette to explain. It came with a 283 hp block, but he has tinkered with it ever since by putting all sorts of bells and whistles on the engine. Now he gets somewhere around 570 hp out of it, but it is still the same 283 block.
So it is with solid cores, you can get more power out of them by adding various bells and whistles, and I want all those who hold to the heavy, fat reactor or lousy T/W ratios arguments to pay attention now or Iím going to give you a whack in the back of the head, just like I got in grammar school to get me to pay attention. The fat, heavy engine and T/W ratio arguments are bogus, a sham: they take a chemical engineís weight and compare it with that of a nuclear engine. Well, of course, the chemical one is lighter; thatís a no brainer. Uranium is one of the heaviest elements. But it overlooks the 90% weight of the fuel and oxidizer, two tanks and engine compared with one LH2 tank and engine and it ignores a rocketís most critical parts: its payload fraction and speed. Itís what weíre spending all this time, money and effort on in the first place.
T/W ratios may be useful when comparing a chemical rocket engine with another or with an airplane engine with another, but here itís diesel/electric sub thinking applied to nuke subs. It doesnít work; for example, with subs, the most common thing is both sail submerged, but one has to surface every couple of days to recharge the batteries while the other can sail submerged around the world and under the North Pole. Big difference! Hereís another example. A B-52 bomber loaded with conventional bombs has a devastating effect in a bombing run, but to load the same B-52 with nukes and go on the same bombing run would be utterly asinine and probably cause the destruction of the B-52 itself. So you need different thinking and new criteria for a new technology. When considering nuclear rockets, I argue the focus must be on the payload put in LEO or sent elsewhere and at an increased speed. Itís all about payload fractions and speed. If I may paraphrase Confederate General Nathan Bedford Forestís military maxim: he who wins the battle ďgets there firstest with the mostest.Ē This is what nuclear does, it gets there ďfirstestĒ (speed) with the ďmostestĒ (payload). So T/W=BS. Itís just chemical rocket thinking.
Iíve pumped up the power density balloon; itís time to let a little air out of it. When you increase power density, you must increase the rest of the system to handle it. This can add weight or other complications, so itís not as clear-cut and straightforward as it seems. Iíll take Stanís Vet to explain. If you rev up a 570 hp engine, then pop the clutch, you may shatter the clutch plate or snap an axle. So they have to be beefed up to handle the increased torque coming from the increased horsepower. The bottom line though is increased power densities and therefore increased thrusting powers are likely as you progress from one-generation to another. So a dynamic mindset is required.
Nuclear and non-nuclear improvements. Other than using the thermal energy from the fission process, there remains the possibility of using the radiation emanating from it for shock heating, radiation heating and so forth, like the radiation coming off a fission bomb is channeled to trigger a fusion reaction, also known as the H-bomb. I know little about this except that some people much brighter than I think you might get 1600 seconds by using it. Iíll just lay this card down now and have a bit more to say about it later.
Many non-nuclear improvements can be made, aside from such obvious things such as making the LH2 tankís insulation lighter yet just as effective. This can save hundreds or thousands of pounds. Another possibility exists - slush hydrogen (SH2) - LH2 cooled several degrees colder than -259įC to where it turns into an icy-slushy mixture, like a snow cone at your local 7-11. Actually, triple point hydrogen is better than SH2, but slush is more memorable to say than triple point, which sounds like a basketball score.
Rover/NERVA was just starting to investigate SH2 when it was cancelled because it shrinks in volume up to ~20%. So if a tank weighed 10,000-pounds to hold a given amount of LH2, it could be reduced to ~8000-pounds for the same amount of hydrogen as SH2. Thus, ~2000-pounds no longer is in the deadweight column but can be shifted over to payload column and so your economics change, because youíre carrying more payload to LEO. The principal concern with SH2, as I understand it, is whether an ice bridge would form over the turbopumpís inlet end, thus blocking the flow and causing the pump to cavitate. After Rover/NERVA ended, Iím uncertain if anyone continued work on SH2; my source doesnít think so, but heís been retired for years and may not be conversant with all going on. This would be a prime R&D goal for a program of the 2010s.
This is a good point to stop my discussion and answer the critic who said my 20,000-pound weight for a LH2 tank in The Nuclear Rocket was excessive. He is right and I knew it was when writing the book, but worried about how the tank would handle the stresses and strains of being discharged horizontally at 500mph from a C5-A and then shifting to a near vertical position. To do that with a solid fuel Minuteman is one thing; to do it with a big and bulky LH2 filled tank is another. To my knowledge, no one has ever discharged a big and bulky LH2 tank from a cargo plane. I donít think itís impossible, just no one ever saw the need to do it. So I beefed it up and am glad he caught it, but I am not worried about it. My entire book and its technical aspects are conceptual on a flight profile not studied since 1960 and are designed to provoke discussion and dialogue. I hoped someone would come up with better analysis. Since then Iíve given more thought to it and believe a winged cradle might be the answer. It would house the solid boosters and the nuclear stage would be nestled in it; the solids would fire and when they burned out, the stage would separate and fly to LEO while the cradle could glide or parachute back for reuse. Conceptually, the cradle would absorb the stresses and strains, allowing the tank to be as lightweight as possible. Iím sure others will come up with ideas here.
Now Iíve come to the heart of the matter and want you, the reader, to become viscerally involved so you get to feel the dynamic quality of nuclear rockets in your bones. I will do this by asking you to design engines and stages and make decisions on what you think can or cannot happen. I hope all read Appendix B of The Nuclear Rocket for insight, as there I laid out a likely progression from a first to fourth generation engine based on the Pewee. I did it specifically to avoid a ďmission-itusĒ mindset that gloms on to one nuclear rocket design for a mission and analyses it to death, seeing the tree in splendid detail, but failing to see the others in the near term development pipeline - failing to see the forest. If you havenít read Appendix B yet, the previous discussion should help. Here I will continue that thinking through a tenth generation system and by the end of this discussion, I hope all will have nuclear rockets in your bones.
So I start with a fourth generation engine with an Isp of 1000, weight ~6000 pounds, and 40,000-pounds of thrust, or 800MW. (This would be a power density increase of 60% from a first generationís system of 25,000 pounds of thrust or 500MW). Estimating the LH2 requirements for all upgrades discussed subsequently would be difficult if not impossible without additional technical specifications, so just assume it will be 50-pounds of LH2/second. Itís in the ballpark for this conceptual exercise. For most Rover/NERVA tests, the norm was around 70-95-pounds/second for reactors which were over 1000+MW, except for the Pewee, which was about half of that at 500MW. The 55-inch Phoebus 2A, still the worldís most powerful reactor at 5000 MW, needed hydrogen at 280-pounds/second. Moreover, donít worry about the technical feasibility of the nozzle and turbo-machinery; they can be developed to handle any Isp and power level you come up with, except for the later generation systems. Remember weíre not designing actual engines and stages here, but trying to get the dynamic potential of nuclear rockets into our bones. So itís conceptual thinking to spur our vision and creativity to develop much better numbers. For the sake of argument, letís assume these improvements occur sequentially; whether they would in the real world is not important here. So our fifth generation system might be introduced ten years after the program restarted.
For your calculations, LH2 weighs 4.43-pounds/cubic foot and there are just under 7.5 US gallons in a cubic foot, so one gallon of LH2 weighs .59-pounds or about a half a pound per gallon. The latter is a good round number for this exercise.
So, as we start your first task, since some held my tank weight was excessive, is to develop your own and from that to develop payload numbers for a fourth generation engine, firing somewhere about 100,000-feet and going to LEO, defined as ~200-miles up. The flight should last about 10-minutes.
Fifth Generation System - Letís assume graphite fuels, after 10-years of R&D, can operate above 3000įC+ and go with a modest gain, say 3100-3200įC, to allow the Isp to increase to ~1025. Some may disagree and say that since graphite sublimates at 3700įC, this is impossible, you have reached the end of the line with graphites. Thatís a legitimate position because somewhere there is a limit to graphite fuels. You make the call. Those who agree should adjust their payload numbers up and those who donít, donít.
Sixth Generation System - Letís assume SH2 is workable, so now we reduce the tank weight and shift the pounds-saved into the payload column. Since SH2 shrinks up to 20% in volume, letís assume three scenarios Ė the ultra-conservative, the moderately conservative, and the optimistic Ė where the tank weight is reduced only 5%, or 10% or 15% for a given volume of hydrogen. Bear in mind here that SH2 is colder than LH2 so the tankís insulation might have to be beefed up, which will add weight, but then again, the insulation for LH2 might be fine for SH2. I have no insight here. So whatever figure you have as your tank weight, then deduct 5%, 10% or 15% of that and put it into the payload column. Then calculate your payload fraction.
Seventh Generation System - Now weíre back at the engine and letís assume the all-loaded core is feasible. Here I have to do a tutorial on graphite. In 1955, Los Alamos picked graphite to be the core structural material because it had good neutronic properties, was cheap and easy to work with and got stronger as the temperature became higher, until it sublimated. However, that strength was only in compression, when pulled, when in tension, it was weak and brittle. Hence the B-4 evolved, with its hexagonal fuel element bundles Ė one in the center surrounded by six. This center element had initially a tie rod ran the length of the fuel element and screwed the entire bundle into a package. In other words, this put the fuel bundle in compression. Think of a nut and bolt that screw things into a package, like the lug nuts holding the wheels to your carís axle or two nuts holding the axle of your bike to the frame. It worked quite well and later tie tubes came along to recapture some of the performance loss Ė since this center fuel element did not have any uranium at all in it. It was strictly for support. Obviously then, this meant a loss of power, as every 7th fuel element was a dud or in technicalspeak, inert.
Then the unexpected occurred, as usually happens in R&D. Since 1955, the fuel fabrication people made better and better graphites and by the early 1970s had come up with varieties that had both good comprehension as well as tension properties - you could pull or twist hard on them and they wouldnít break. (You know this if you have a golf club, tennis racket or fishing rod made out of graphite or carbon composites. All this spun off from the pioneering work Rover/NERVA conducted on graphite). This promised an end to the 7th dud element and it doesnít take a rocket scientist to see that by going to the all-loaded core, where every dud now has uranium in it and is fissioning, you can increase your power level by 7th. So your power density is going up without doing any of the other tricks to get a higher power density. Just replace the duds with live fuel elements. Unfortunately, the program ended before Y-12 could start work on this.
(Another way of going to the all-loaded core might be to adopt the dome concept developed for the Pluto nuclear ramjet. Domes are exceedingly strong and to prove it, take a fresh egg from your refrigerator, clasp your fingers together as if praying and place the egg so one of its ends rests in one palm and the other in the other. Then squeeze with all your might. Donít worry it wonít break except if you are exceptionally strong. Now visualize cutting the egg in half, so it forms a dome. The interior part would be the coreís hot end and the edges would attach to the pressure vessel and here you would have to pay special attention to cooling them. In 1961, Pluto ran for five minutes at 500MW and almost 1400įC and produced 35,000 pounds of thrust. This domed core concept further illustrates the dynamism of nuclear rocketry).
When we look at our small engine, we see it had 534 fuel elements, but 132 were duds, much more than the KIWI, Phoebus and NRX reactors. This is about 25% of the core, rather than a 7th. The reason was neutronic, all duds were coated with zirconium hydride, a good neutron moderator, to promote fissioning because the 20-inch reactor had a problem going critical. Now, however, were increasing our power by a quarter without adding any of the other power density tricks and if we add that to our 40,000-pounds of thrust of our fourth generation system, weíre getting close to 50,000-pounds of thrust. So this is just about a doubling of power from when we started the program with our first generation system with a thrust of 25,000-pounds, so we need more hydrogen. The weight of the entire stage will probably have to be strengthened to handle this, so that will increase its weight somewhat.
Now calculate the payload, and start with a nice round number thrust level of 50,000-pounds, then back off a bit if you like. Whatever number you come up with, it should be quite clear by now that the payload fractions are really increasing. Theyíre dynamic.
Eighth Generation System - Over the years afterburners attracted a lot of interest, though what constitutes it is open to debate. One definition might mean doing something to the hot hydrogen plume as it leaves the nozzle throat and enters the divergent section of the nozzle. There are different ideas on how to do it, some of the older ones being radiation or shock heating while some of the newer ones being to use a laser to add energy to the propellant. I donít have a great deal of knowledge here, so Iíll just leave this aside and take it up briefly below.
Another is the LANTR concept of injecting oxygen into exhaust plume after itís left the nozzle throat and the Glenn Research Center has pioneered this thinking. Here oxygen would be injected into the hydrogen plume and burn, giving an afterburner effect; but if that was not desired, the engine could run on hydrogen alone. LANTR would require major modifications to a simple nuclear engine, and itís something to be considered by a more mature program, but it just shows the dynamism of nuclear rocket technology.
One technology, however, has been viewed as an afterburner, but probably isnít since it doesnít involve doing something with the exhaust plume after it leaves the nozzle throat. That is the graphite/carbide hybrid fuel element. Since carbide fuels promise operating temperatures well above 3000įC, they likewise promise much higher specific impulses. In its final years, Los Alamos was moving more and more to carbides and if the program had continued, it would have shifted over to them exclusively. They really liked their potential of 1100+ seconds of Isp. As noted, however, carbides are brittle and would need a new core design, the B-4 wouldnít do.
But thatís not as black and white as it sounds, because there were concepts to cut part of a graphite fuel element off and braze a carbide element to it. How can that be? Brazing surely seems not to be something we want to do in a core operating at 3000įC+. Let me explain. The 52-inch long fuel element is not a uniform temperature, but it increases gradually from 0įC at the cold inlet end to its hot end where it would be 3000įC+. So all you do is determine the melt value of your braze material Ė say 1500įC - then go to that spot on the fuel element where it is that temperature (or just above it), cut it off and braze a carbide element long enough in its place to make the fuel element a full 52-inches. It sounds simple, but to my knowledge, there was never any R&D on it, perhaps because carbides appeared late in the programís history, when its death was imminent. You also must do something about supporting it, as it couldnít just dangle, so the Pluto domed concept discussed earlier might come into play.
Could it work? I donít know. If yes, it might mean an Isp ~1100. So hereís another judgment call. Those who think yes should analyze at 1100 and get their payload number and payload fraction. Those who think no should pass this by. All, however, should note the afterburner card is still on the table and Iíll talk about it next.
Ninth Generation System - This involves disassociation of the hydrogen molecule into two atoms, or call it the use of monatomic hydrogen. Throwing a single hydrogen atom out the nozzle is stunning because now in our rocket equations of T/M, M is now 1 and not 2. This means a big increase in Isp, letís say to 1600 or so seconds. Or more. It all depends on what T is. It also means a big change in the space program. Well, is it possible and if so, why havenít we done it?
Here is my view, which should be taken with a grain of salt because Iím uncertain of the data and because this might involve radiation and shock heating or lasers in the nozzle or elsewhere or something else Ė things Iíve already said I donít fully understand. Technically, one source indicates hydrogen disassociation starts at 2000įC and finishes at about 5000įC. Itís like a bell curve. Another says this occurs at 3000įC to 3300įC, but is so unstable that it recombines in a flash. Then another says radiation weakens the bond of the two atoms to each other so that the disassociation can occur at lower temperatures. This implies the 3000įC to 3300įC level might be accurate for hydrogen in a high radiation environment while the 2000 to 5000įC bell curve might likewise be accurate in a non-radiation environment.
Either, however, says hydrogen disassociation is not theoretical, but needs applied science to make it happen. So why hasnít it? Hereís my take. The hydrogen atom does not like to exist by itself and so recombines to form a hydrogen molecule almost instantaneously, giving up a great deal of energy in the process. If this happens in the nozzle chamber itself, youíve suddenly got a lot more energy to deal with, making an already difficult heat transfer and other problems that much more difficult. To avoid this with LH2/LOX, they run the engine slightly rich in LH2, so it doesnít burn with LOX and exits the nozzle throat as diatomic hydrogen. I think this is accurate. With nuclear rockets, they also donít want the monatomic hydrogen to recombine in the nozzle chamber, that area between the end of the fuel elements and the start of the nozzle throat; it would make really difficult heat transfer and materials problems that much more difficult. (This is not to be taken lightly; the tubes used to cool the convergent portion of the nozzle were 7-mils thick Ė this is real thin - to permit heat transfer, so shocking them with even more heat presents real structural integrity problems). So they want to wait till the diatomic hydrogen has left the throat and is in the divergent portion of the nozzle and then do something like shock or radiation or laser heating to make it disassociate there and there deal with its problems. So itís a true afterburner. I think this is right.
If Iím accurate, and I might not be, why didnít Rover/NERVA work on it? They certainly knew about it and wrote occasional papers and Los Alamos thought about making an afterburner R&D a formal effort in 1967, but I think they had yet to get to the right temperature levels. Most tests were around 2000įC and only in the final three or four years of the program did the temperatures start to rise Ė Pewee operating at 2265įC in 1968 - and they were starting work on graphites operating at 2600-2800įC and moving into high temperature carbides when the program was cancelled. Moreover, Los Alamos was just starting to investigate lasers at this time and some early papers thought of laser fusion propulsion, the lasers imploding pellets in a chamber, with the resultant debris and energy expelled out a nozzle. This was sort of a miniature Orion nuclear bomb propulsion scheme. So from their perspective, afterburners were premature, not impossible.
I believe todayís program would focus on this, because 1600s seconds, or more, would be dramatic and change not only the economics of getting to LEO but also the movement of payloads from LEO outwards. This is a game-changer. But is it feasible? Thatís what R&D would attempt to figure out. Many ideas exist now and I believe after teams of people start working on them even more would be found. It promises to be a very fruitful area. How long would it take to sift through these concepts and wild blue-sky thoughts and find something that actually works? I donít know, the problems are quite formidable, but to view it as something that might happen after a decade or so of R&D may be the right mindset. Teams of people who work on and test concepts experimentally, not just theoretically, have better results than the lone researcher or two with their pens and papers and computers. And remember the NERVA I that existed in 1971, and the missions it would enable, was not even dreamed of in 1960, so radically different was it. To spur work on afterburners, make it an X-Prize. It really is a game-changer.
Tenth Generation or the Serendipity System R&D is a process that at its beginning the goal is to gain knowledge about something one knows little or nothing about - you want to learn what you donít know. As you continue, R&D changes its focus and shifts into applied science. Here youíre trying to make something work - proving what you do know will work to high degrees of reliability. With NERVA, it was to be 0.997 reliability. In this process, you invariably find out things that you never dreamed of before hand and so we might call this the Serendipity Generation (serendipity means finding valuable or useful things not originally sought after, like going fishing and catching one that swallowed a diamond ring). Thus, it ties in nicely with what I discussed about hydrogen disassociation. Here itís right to expect serendipity different from what Iíve just discussed. Iíll say it differently: I expect improvements to the B-4 core beyond what I have discussed here. Moreover, those core concepts other than the B-4 might prove feasible after a decade of effort. So here, in principle, one might find a very dynamic development effort. I would include the Vasimir nuclear rocket concept here as well as other nuclear ion systems.
Going back to the B-4, however, one class of examples might be using ammonia, methane or water as the working fluid. These have molecular weights around 18, so why even consider them when you have LH2/LOX? Well, the answer is simple, when operating ~3000įC, NH3, CH4 and H2O disassociate into atoms; so M is around 8. This promises Isp around 600 seconds or so. Thatís not as good as hydrogen at 2, but itís better than 450 and for some missions it might be just fine. Krafft Ehricke, the famous V-2 German who later became a key figure in the US space program, conceived of a two-stage nuclear rocket, the first being a piloted nuclear ammonia stage carrying a LH2 nuclear stage to 150,000-feet whereupon they separated, the latter going to LEO while the former flew back. Suffice it to say this might leave the pilot very sick or dead, but with high Isp and high power density engines it might be possible to devise radiation protection for him and still have the capacity to haul a large second stage to 150,000-feet. No one, to my knowledge, has looked at this since and on the face of it still seems pretty unappealing, but who knows if high Isp and power density engines change that assessment by allowing extra shielding for the pilot? Moreover, drone technology of today could eliminate the pilot altogether, making an ammonia first stage an intriguing concept. Who knows though? If you never study it, how can you say it will or wonít work? In sum, a major benefit from using these fluids is that they are denser than LH2/SH2, so the tank can be much smaller, saving a lot of weight. It dynamically opens up design choices.
Another Serendipity category is the carbides, of which there are many types that promise higher temperatures than graphite, but are brittle. Back in 1958, you had the tungsten Dumbo concept, which many talk about today. It featured laminar flow heating, in which the fuel element would be 0.425 inches long, with a propellant passageway 0.0055 inches wide and an inlet width of 0.00126. A 0.001 inch-thick coating of uranium would cover these metal strips. Thousands of these elements would be stacked together and form tubes that in turn would be stacked to comprise the core. These extremely tight tolerances meant the hydrogen flow would be severely constricted, like soldiers marching in a parade, and be heated to very high temperatures in a very short distance. In principle, Dumbo promised high Isp but very low weight engines. (Rover/NERVA had turbulent flow, where the hydrogen molecule bounced off the sides of the 52-inch long propellant hole as it picked up heat). It had two critical problems: tungsten was hard to machine and weíre talking about thousands of fuel elements to make a core, each the same thickness of the other, so you have obvious QA and QC problems to deal with. Second, the uranium had to be vapor-deposited on each of the fuel elements in a uniform thickness, otherwise you would get hot and cold spots. Here was another QA and QC problem. It was cancelled in 1959. Today, however, machine tools are dramatically better than back then and vapor-deposition is a well known technology, not a developing one, so the nuclear rocket program of the 2010s would be right in reinvestigating Dumbo.
Thatís not all with carbide; there are other core concepts. The Russian nuclear rocket program mysteriously switched from graphite to carbide in 1968 and spent the ensuing decades testing different carbides. They proposed a core concept of taking very thin, long carbide strips, depositing a uniform coat of uranium on it, then taking a bunch and twisting them, as if one took a handful of drinking straws and twisted them. Then bundles of them would be added to form the core. Itís an ingenious way to add structural strength but it still remains to be seen how the core would stay in the pressure vessel. Despite published statements, the Russians have never flown or flight-tested it nor, to my knowledge, tested anything other than fuels. So who knows if this core would stay in the pressure vessel? Iíll conclude my summary on carbides by saying as they are investigated many core concepts would evolve; at some point that serendipity moment might occur.
In sum, I think by now all should see the sheer dynamism of nuclear rockets. You should have an intuitive feeling, you should have it ďin your bones.Ē Or at least I hope you do. Contrast that with a chemically propelled one where theyíre trying to get a tiny improvement in payload fractions. Despite billions and billions and billions being spent since the 1960s, how larger have payload fractions become? Donít believe me, do the exercise yourself, that way you will be convinced. Put the payload fractions of the Atlas, Titan and Saturn V of a half century ago in a row and then add those of the improved Atlas, Titan, space shuttle and Ares of today. You will find a fraction of a percentage point improvement. Itís moribund and will be so as long as the T/M govern how we access space. So we need to go back to basics and focus on T/M - increasing T, decreasing M. And if I collected your tally sheets from doing the exercises, I think I would find a steady upward progression in payload fractions. Nuclearís dynamic. And I think you will also see this sheer dynamism requires a mindset is not dominated by ďmission-itusĒ thinking or one that aims at developing a nuclear rocket for manned Mars.
These are not hard numbers, but then you donít need them now. You need vision, faith and imagination; as you refine concepts, better numbers will come. But remember the best numbers only come from actual tests, particularly those at the tail end of the R&D cycle, as here the focus is trying to get something for flight-rated status. Those numbers must be hard as payloads and perhaps people are involved. By now, I hope some have taken a step in my direction, but that may be a hesitant step and most really remain back with the others who are now chanting ďWhat about safety?Ē OK, letís talk about safety.
Safety can be divided into three categories: workers, the public and the environment. In Rover/NERVA, three workers died and the cause was carelessness or ignorance. Those would kill the workers of a renewed program, just as it does in any industry or household now. Dumb is dumb. For more information read my appendix on Safety and the Environment in To the End of the Solar System. Next, workers were exposed to radiation and here I can only detail what appears in the public record as health records of individual workers are protected against public disclosure. So at best I have anecdotal evidence that exposures were below the guidelines though there were instances of some getting higher doses. Stan Gunn, for example, picked up the pieces following the TNT test in which a full size 1000MW reactor was purposely blown up (read about that in my first book) and says he got 10R from that alone, which is double the current annual limit. Heís 87 now and in good health and I can attest, likes a couple of drinks with dinner. I firmly believe is radiation is vastly over-hyped or a different way of saying it is that the human body is really quite remarkable and can repair the effects of a lot of harmful things, including radiation exposure. Moreover, all should know that space is intensely radioactive and going there will mean exposures much higher than on earth where our atmosphere blocks out much of the radiation. The bottom line here is radiation protection for workers is a non-issue; itís been done well for decades from your dentist and hospital to nuclear power plants and other nuclear endeavors, such as the nuclear Navy.
On testing, no reactor was shielded and all gave off gamma rays and neutrons directly to the environment. However, air absorbs this in 3/5s of a mile. If you look at the test complex at Jackass Flats, you will see the control room building is a mile away from the test cells. No problems here for a renewed program. Chemical rocket testing also features an exclusion zone. All tests except one vented a radioactive plume directly to the environment, and again read my appendix on Safety and the Environment for how they monitored and took steps to ensure public safety hundreds of miles away from the Test Site. You will also read the EPA conducted a wrap-up study of all tests in the program and gave NASA and AEC kudos for a well run program, one whose effluents were under the guidelines, as I mentioned earlier. The program of the 2010s, however, will not vent a radioactive plume directly to the environment and the why is simple.
In 1971, the Nuclear Furnace (a real small fuel test reactor) had a scrubber to remove the radioactive particles from the plume before discharging it to the environment. In concept, it is a large container that fits over the nozzle and through which high-pressure water is pumped. This cools the hot plume and condenses it into water that is then filtered to trap and concentrate the radioactive particles. After further processing the scrubbed-clean gases are vented to the environment while the concentrate disposed of accordingly. In the Nuclear Furnace test, the EPA found trace particles of radioactivity several miles from Test Cell C, but none off the Test Site. (A trace particle does not mean it is dangerous or harmful, only that our instruments are so sensitive now that they can detect different particles down to a few parts per billion). That was a first generation system; better systems were planned but the program was cancelled. A program of the 2010s would have scrubbers. That takes the issue of public and environmental harm from testing off the table.
Now we come to the final issue: launching a nuclear rocket to LEO and returning it from LEO. Can we do that safely? Iíve partially discussed the flight profile here but in Appendix C of The Nuclear Rocket I have a ten-step walk-through to see what the major problems are in each step and how they might be solved. I urge reading that. I donít see any show stoppers, but you form your own opinion. Here, however, let me hit some of the highlights.
First the flight profile starts in a very isolated ocean area where few, if any live. So no people, no public risk. Second, the engine starts above much of the atmosphere or over 100,000-feet and goes to LEO. So worries about radioactivity in the air we breathe from a nuclear rocket is nonsensical; if one wanted to worry about that, then worry about radon which is present in many of our homes or buildings, particularly those that have a lot of granite in them. Now each engine would not be fired naked, with the core surrounded just by a metal pressure vessel an inch or so thick as in Rover/NERVA, but it would be encased in what I call a cocoon, though you could call it a wrapper, enclosure or something else. This is new and to the best of my knowledge, never studied earlier.
It would allow the nuclear engine to return from LEO to the earthís surface where it would be recovered with minimal risk. Four different shapes appear possible: an RV type of body (that protect nuclear weapons during reentry), an Apollo-type mushroom, a lifting body shape similar to the space shuttle orbiter or a clam-shell apparatus.
Some might feel uneasy over this, but these systems have been proven over the years and can be quite rugged. Iíll give just two examples. First, I had a senior moment in writing The Nuclear Rocket and forgot a key point to demonstrate it; those interested can go to my book entry on Amazon and read my ďMea CulpaĒ comment. In short, it involves the Mark-6 RV for the Titan II. It held the W-53, a 9.2MT warhead. In 1980, a Titan II with a Mark-6/W-53 blew up in its silo. The force of the explosion hurled a 740-ton launch duct door, which covered the silo, 200-feet in the air and it landed 600-feet from the silo. The Mark-6 followed, bounced off the door and landed several hundred feet from the silo. Neither the W-53 nor its high explosives went off. The Mark-6 was recovered battered but intact. No release of radioactive materials occurred, no plutonium or uranium was scattered over the environment. Everything stayed either inside the weapons casing or within the Mark-6.
Second, the early SNAPs, which used radioactive materials such as plutonium-238 to produce electricity for payloads, were designed to burn-up completely if they reentered the atmosphere and in 1964 that happened. But it prompted a change in design philosophy that henceforth emphasized maintaining the integrity of the radioactive source in all aspects of the flight profile. In 1969, a launch was aborted from California and two radioactive sources fell into the ocean and were recovered intact shortly thereafter, with no impact on the marine environment. I have many other examples, but I contend if we can build a system that can withstand the violence of being blasted out of a silo or suffering a launch abort, we have the technological muscle to design a cocoon that could withstand all plausible accident and recovery scenarios of the flight profile to and from LEO. Itís time for our scientific and engineering community to stop doubting its prowess and to start considering this flight profile.
Moreover, itís easy to do and will cause no environmental harm whatsoever. Much of the data on reentry shapes is in the public domain and so we can make use of it in modeling, to see how such a cocoon would behave. We risk nothing by these computer studies and once weíre convinced of a prototype, we can test it on rockets, but the cocoon need not hold a radioactive core, just something like a tungsten mockup (tungsten and uranium weight about the same) and if we wanted to see if it leaked any radioactivity, we could even put some hospital isotopes in it, the kind they inject into your arm to detect illness or disease. How dangerous could that be if they leaked? This turns the R&D into a public relations stunt, but in todayís environment that is often necessary.
While those reading this introduction could do the modeling, they could not design the internals of a cocoon or they could do so only with the help of nuclear and space engineering departments at universities or with the unofficial support of NASA or DOE. Inside the cocoon would be all sorts of gear to prevent or mitigate any type of accident such as helium (used in Rover/NERVA to cool a core down quickly), special foams to harden in the core to prevent accidental criticality, parachutes and floatation gear, black box recorders, telecommunication equipment, and the list could go on. There could be a lot of room to work with, depending on the size and shape of the cocoon. Moreover, the rest of the cocoon could be filled with neutrons and gamma ray absorbing materials such as boron or even water and thus dampen the radioactivity while the engine remains in LEO and limit it for when it is picked up after it returned to the earthís surface. Itís deadweight, which detracts from an improving payload fraction, but thatís just the cost of doing business. If 1600 seconds were possible, this would just be a blip.
Obviously, this would be a dynamic area, with many ideas possible, such as eliminating the pressure vessel to let the cocoon proper support the core. But it also has problems, such as rejecting the heat or channeling it via heat pipes that would build up in the cocoon. Nonetheless, these designs could be modeled and the most promising tested at Jackass Flats or elsewhere with minimal risk. We would be dealing in this phase with mockup cores and perhaps hospital isotopes in various shake, rattle and roll tests. Thereís no risk here. At some point, they must be tested live, but letís not get paranoid and worry excessively over the marine life. The oceans were even more radioactive millions of years ago than today, and the marine life survived, the atmospheric nuclear weapons testing era added radiation to the oceans, and the marine life survived and even thrived Ė we added Godzilla - and cosmic radiation creates isotopes in earthís atmosphere daily that fall into the oceans, and the marine life still survives. Moreover, several nuke subs unfortunately have been lost over the decades, without noticeable harmful effects on marine life. Finally, the Soviet Union crapped up its fresh and salt waters with radioactivity real bad, but the marine life still seems to have survived. Thus, this seems to be more an emotional issue than a technical one. Yet when this time for hot testing comes, it will probably be more like a demonstration, so even here risks should be quite low. In other words, Rover/NERVA pioneered the use of modeling techniques so when a hot test was conducted, it was more a ďdemonstrationĒ of the model, than an ďexperimentĒ to learn something new, to see if something works. Once that begins, I have every confidence in our technological capabilities Ė recall the Mark 6 that was blasted out of a silo and didnít spread radioactive materials to the environment - and hold the ďriskĒ would be acceptable - quite low on a probability scale. I think it would be like those we take when we fly from NY to LA.
I hope by now I have persuaded all to my point of view, but some might be saying: ďOK, OK, we believe you. But they will never let us do that!Ē I have three comments to that.
First, you now have new justifications for todayís program utterly different from manned Mars. Theyíre contained in the second part of the title of my second book: Making Our World Green, Peaceful and Prosperous. These are very powerful and all need to think deeply about them. They are a paradigm shift. When they say it will pollute the planet, you can say ďNo, No, No, you got it all wrong. It greens it by allowing us to ship environmentally toxic or strategically dangerous materials there initially and later to move our industries there. Why keep the earth forever constipated by allowing a continual buildup of these materials here? Donít bury them underground, where they contaminate our ground water or dump them at sea and where they pollute our oceans, but send them into the vastness of space. We are the true greens!Ē If they say solve global warming first before going into space, you can say ďAh, Yes! The low cost of reaching LEO puts all space-based options on the table for solving it, not just earth-based ones. If global warming is as serious as you say, then you must put everything on the table to solve it. If you donít youíre a hypocrite and not serious. So join us, we are serious about it.Ē If they seek peace on this planet, you can declare ďRight On! Some of the most brilliant scientists of the 20th century said nuclear rockets would bring peace to Earth. If youíre serious here, then you must join us and put all options on the table for achieving it, not just some.Ē If they say space is too expensive and we should solve hunger and poverty on earth first, you can say ďPrecisely, a nuclear rocket space program will be democratic as well as prosperous, so weíre looking at ways for all to participate and profit. To ignore that is to ignore all the economic benefits the nuclear and space programs have brought to the worldís economy, such as the Internet on which you are currently reading this note. Here we can find ways to let those in the most humble of circumstances and most out of the way places participate and profit.Ē To all of this, you can say ďWhy donít you join us instead of working to keep the world dirty, warming and warring? Give peace a chance!Ē
Second, why do you accept second-class status? Why do you assume the anti-nukes, the greens and the peace-niks would prevent you from doing it? Why do you give them a veto when in a democracy all votes should be equal? Doesnít your vote count? Do they have all-wisdom and all-knowledge in how to protect the earth and you, who are technologists to one degree or another, are lumped into that category of ignorant and dangerous polluters? Nonsense! And why assume politicians would oppose when breaking the ban promises a revitalized and democratic space program open to the common man, bringing in its wake many high-paying jobs and industries and an expanded tax base? Politicians may not know Isp and payload fractions, but they know full well the common man votes and they also know full well the importance of jobs, industries and tax-bases. I insist they will listen keenly when proposals are made to restart and run the program with private money.
Three, breaking the ban promises a revitalized space program funded mostly by private money, not government money. It unshackles the creative and innovative and competitive nature of the common man who will in turn form companies to do things in space other than national security, commercial satellites, or science, or to do the latter three better and cheaper and I will add, faster. Thus it achieves NASAís mantra of ďfaster, better, cheaper.Ē So look for the creation of many high paying high tech jobs, particularly when heavy lifters appear and real large space stations follow, not the fly-weight 440-ton ISS. If you doubt this, again remember the Saturn-NERVA that could push up to 250-tons to LEO. This transforms space from being very low in the public opinion polls to one very high, as it was in the late 1950s and early 1960s. It also can revitalize the nuclear power program in the US and create many jobs at key DOE facilities and at the same time allow the US to have a powerful diplomatic tool to achieve nonproliferation and other foreign policy objectives. I canít discuss all these things here, for that part of the argument please read The Nuclear Rocket, but a good summary for here is jobs, new industries, expanded tax bases, money and power. Politicians will listen when you talk those five things.
So breaking the taboo begets a dynamic program with an equally dynamic set of new arguments to appeal to the anti-nukes, the greens, the peace-niks and the politicians and the public. Donít view them as hostile, I certainly donít, but donít accept second-class status, I certainly donít. Approach them with a new way of thinking about space and how the low cost of reaching LEO by breaking the ban can help them achieve their goals. Get them on your side with new thinking. They may turn out to be the programís best supporters. Whether they still oppose or not, go to the common man and show him or her how to participate and prosper in this new space program. Get them on your side and it should be easy to do - jobs are a very powerful argument - then let the votes be counted. This is a democracy after all. See how completely different that is than trying to justify government funded manned Mars or a lunar base. Both are politically dead. How can you justify them when the poverty rate in the US is increasing dramatically?
Implicit in all of this is lowering the costs of taking payloads to space for $100/pound, which has been touted as the holy grail for making space pay for itself. Itís a claim made on the back cover jacket of The Nuclear Rocket. Can I prove it and not just assert it? Yes. There are seven factors that will combine to make that figure a reality. Iíve already discussed three: Isp, power density, and nuclear and non-nuclear improvements. By now everyone should see how dramatic are the increases in payload fractions and payloads numbers, so I need not repeat these. However, there are four others.
Before beginning, it is important to go the NERVA schematic diagram (itís in both books) and have you, the reader, do a part-by-part cost estimate of each. This exercise will also show just how simple a nuclear rocket engine is and by implication, why 0.997 reliability can be achieved. Simply stated, the more moving parts a system has the less reliable it becomes. Thatís a good general rule, but nuclear adds its own complications as its radiation can cause damaging changes in a material, so itís not as clear-cut as it seems. Still, that is something that can be fairly well estimated in a good materials testing and qualification program. With that caveat, a nuclear rocket has the following components:
I cannot include any costs for the cocoon, as it has yet to receive even a conceptual design. So this is an unknown. I also tried unsuccessfully to get the cost of the solid boosters, so this is another unknown though the readers may good data here. And I have no data on a cargo plane such as the C-5A.
In The Nuclear Rocket I estimated the cost of each engine at $25 million, which I think is high, but itís good enough for conceptual thinking. Here, however, I want the reader to become involved and make his or her own estimates. How much do you think a pressure vessel would cost? $50 million? $5 million? $500,000? $50,000? I donít know the price of boron but beryllium is about $250/pound. A 100-pounds of boron and 1000-pounds of beryllium might be high, but there will be fabrication losses for both and beryllium requires special handling, as it is a health hazard. All major non-nuclear components such as the nozzle and turbopump/turbine are 1960s era technology, so their costs should be low. By now, the reader should realize intuitively the costs are going not to be high Ė even if he or she hasnít done a cost estimate - so any talk of nuclear rocket engines costing hundred of millions or billions is absurd.
Now let me discuss the four other factors that will drive costs downward.
I must now focus on the core itself and give some background on fuel fabrication. In three sentences, to make fuel elements, you take a large commercial kitchen sized mixer and pour in graphite and HEU, mix until it becomes a slurry, sort of like bread dough. Then run through a high-pressure press to make it denser, and then through an extrusion press to shape it into fuel element form. Then it cover it with various coatings to protect it from hydrogen. Iíve left out a lot and exaggerated quite a bit, but still this is a good summary. Saying it differently, fuel element R&D is painstakingly hard and difficult, but once you have the right ďformula,Ē itís rather straightforward in production. For further information, read my appendix on fuel fabrication in To the End of the Solar System.
In Rover/NERVA, Y-12 stated each fuel element cost $1000 to make and in The Nuclear Rocket Iíve normalized that for inflation. So I said each fuel element would cost $4000 and thus a Pewee-sized core with 532 fuel elements would cost about $2.2 million or about $2.5 million as a round figure. Letís take a closer look at this figure though. Y-12 was making fuel elements for the 10-hour/60-recycle goal and this was extremely hard. Each fuel element received 2000 QA and QC inspections, thatís right 2000, meaning it was quite labor intensive. For a 10-minute flight to LEO, the fuel element objectives, I submit, would be quite different and emphasize high Isp and retention of fission products (making fuel that would not leach out radioactive particles).
This will be much easier, relatively speaking, than 10/60 fuels. In going from high temperature to cryogenic temperatures and back again repeatedly, cracks form that do not seal up again. So the more recycles, the more cracks and the more uranium loss from hydrogen corrosion and damage to the fuelís integrity. So 10/60 was daunting. In going from ambient temperature to high temperature once and for only 10-minutes, the element will stabilize at one temperature and then is shut down, never to restart. This implies much less cracking and uranium loss, and a much easier fuel production process, one with much fewer QA and QC inspections. So I expect the price to be high for a first generation engine, but to drop as experience is gained. So expect a downward spike, but to what level? For conceptual purposes, try three options: $3000, $2000 and $1000 per element. This means the core would cost about $1.5, $1 and $.5 million respectively.
In going to LEO and returning, the core would be used once. I call this the Re-core engine, i.e., after it is recovered, the pressure vessel is opened, the radioactive core removed and sent to DOE for reprocessing. This is because the engine uses 93% enriched uranium, but in the 10-minute flight, it would use less than 1% of the U-235. So it makes economic sense to recover the remaining uranium and reuse it. DOE would pay for this.
To repeat the Re-core flight profile: it fires to LEO and returns and then itís opened up, the radioactive core removed, and the non-nuclear components such as the reflector, control drums, actuators, pressure vessel, turbopump/turbine and nozzle reused. Actually, they would be recertified, rebuilt if necessary or replaced totally. (Compared to the space shuttle, this should not be a lengthy process). Then a fresh or non-radioactive core is inserted and the launch process starts again.
I estimated these non-nuclear components could be reused 100 times for a 10-minute flight to LEO, 1000 minutes total. How did I get that number? Well, I looked at the 10/60 goal for the first generation NERVA I; 10/60 meant it would run for a total of 600 minutes at full power with 60 stops and starts in space - with no maintenance or repair of any kind. Thatís very hard. So I reasoned if you could operate these non-nuclear components in space for 600 minutes with a total lack of maintenance or repair, 1000 thousand minutes with recertification, rebuilding or replacement on earth seemed reasonable for a conceptual study.
Conceptually then, what are the payoffs? Very low re-launch costs! Perhaps just the cost of the cargo plane, some solid boosters, LH2 and the core, which Iíve just discussed and which could be from $2.5 million to half a million. You make the call. I know it seems absurd, but if the program of 40-years ago could do 10/60 without maintenance, the program of the 2010s should do something just as good if not better.
In the real world, however, todayís program would have a rigorous component testing and qualification program from which they would derive accurate numbers on the life expectancy of each. There might be great variation here, some lasting for 500 minutes while others for 800, even others 1200 and some for only one flight, such as the nozzle extension. And there would be complicating factors that must be dealt with, such as neutron activation of the components; simply stated, this means the non-nuclear components can become radioactive though not as much as the core. And neutron activation might give some handling problems during re-launches. So itís not as clear-cut as it sounds and it certainly requires study, as the payoff can be striking - another downward spike in the cost of taking payloads to LEO.
As experience is gained in taking payloads to LEO, the price per pound will decrease and as it does the law of supply and demand will increase in importance or as the price for a desired service decreases, the demand for the service will increase. Itís Economics 101. This increase will force a shift in thinking from the ďmission-itusĒ approach with chemical rockets where once a mission is defined and approved and then a rocket is built to fleet thinking, the procurement and operation of a number of nuclear rockets. Some will be in the launch queue, some in LEO cooling down for their return to the surface, some in the disassembly and recertification process, some in the preparation for launch and some in ready reserve. Itís what NASA had thought would happen to the space shuttles, but obviously didnít or perhaps couldnít because of their complexity.
When you have fleet thinking, you also get fleet procurement and this will lower costs. Everyone knows if you buy in bulk, you get bigger savings than buying one at a time. For example, letís assume a fleet consisted of ten nuclear rockets. Now ten pressure vessels would be ordered and ten nozzles and ten turbopumps and so forth.
Linked to this is private sector procurement. I argue for creation of NucRocCorp, a government/industry corporation, with industry in control. It would follow private sector procurement practices, which should be economic and efficient, and not those followed by the government. Remember those $1000 Pentagon toilet seats? Or $1000 hammers?
Thus, these two factors, linked as they are, would put downward pressure on the costs not only of taking payloads to LEO but also elsewhere in space. At the moment, though, itís too early to tell how large it would be. The reader though might be able to make some estimates. If 1000-pounds of beryllium is needed per engine and if we have a fleet of 10 engines, thatís 10,000-pounds total. What would be the cost savings in buying that amount rather than 1000-pounds? 10%? 20%? 25%? You decide.
I argue to develop small engines first and after experience is gained, to progress on to larger systems. Some of these will be used to take payloads to LEO. This highlights the versatility of the B-4 core because no insurmountable technical problem exists in going from a 20-inch core to larger ones. It might go in five-inch increments: 25-inches, 30-inches, 35-inches, 40-inches, 45-inches, and 50-inches up to a Phoebus 2A size 55-inches. Obviously, as the core increases in size so do its thrusting power and so more payload can be sent to LEO. Remember that Rocketdyne study that said a 35-inch core NERVA could put up to 500,000-pounds in LEO. The critical thing will be the boost: perhaps for cores under 35-inches, a cargo plane launch would be feasible, but over that other systems might be required: solid rockets, LOX and kerosene, even the Krafft Ehrickeís ammonia first stage. Itís too early to tell. It should be obvious to all by now that this would be a very dynamic area and many different heavy lifters might be built, the same way that there are many different-sized trucks for city deliveries to the 18-wheelers for cross country hauling.
When heavy lifters appear, they will change again the economics of moving large payloads to LEO. In The Nuclear Rocket, I conceived of a 35-inch core capable of lifting 75-tons per launch to LEO. A specially designed cargo plane would carry it. Donít get hung up on the numbers, because its conceptual and its real purpose was to advance the concepts of charters and zones and linking them to the improvements in nuclear rockets. This is quite opposed to manned Mars or lunar base thinking: this says build an extensive, private sector infrastructure in LEO and then move it outwards as successively improved nuclear engines appear. In other words, build lots of really big space stations in LEO for tourism, industry, science, global warming and so forth. They would not be government funded and government run like the ISS, but private sector enterprises. When improved nuclear engines appear, with ever-higher Isps, move the zones ever outward and here economic and profitable lunar bases might be in the offing.
The above centered on cost per pound to LEO, but now we must consider profitability. How much money can be made? Itís fairly obvious that $100/pound to LEO promises to be a very profitable enterprise, even if Iíve excluded the other costs that must be figured into any venture such as plant and equipment, salaries, insurance, health and retirement programs and so forth. But there is another factor, another way to make money and it might be as profitable or even more profitable than building and operating fleets of nuclear rockets. And it is absolutely, positively different from chemical rocket engines, which offer no such possibility. None! Zero! Zilch! Nada!
I speak of technologies that a private corporation such as NucRocCorp would license, sell, or spin-off for a profit. This was not possible in Rover/NERVA because government money was used so the technologies had to be public domain. So the taxpayer got nothing, but individuals or corporations in the know got quite a bit. Here are two examples. Iíve already mentioned graphites and carbon composites and to this we can add heat pipes. Both originated in Rover/NERVA and now are multi-bullion dollar industries. For more information here, read chapter 16 of To the End of the Solar System.
In a renewed nuclear rocket program, there will be, I believe, many profitable spin-offs but not as free-bees, but for some type of remuneration to NucRocCorp and its stockholders. While many possibilities would exist, I can see already two. Both appear to be quite profitable. First, the program will be creating a broad and fundamental assault on high temperature materials that will, in turn, be very marketable and profitable in other industries. Higher temperature materials promise much higher efficiencies, among other things, so you are likely to see market pull here as these other industries see they must incorporate these materials in their products or face loss of market share.
Second, I speak of using nuclear rocket technology for electrical power and process heat applications; take whatís being developed to operate at 3000įC and de-rate it to operate at 1000įC or so. Again, this is not any original thinking on my part. In 1959, Los Alamos did exactly that with the UHTREX reactor project, but it failed after running for 30-days in 1969 and the project was cancelled, purportedly because of a political and not a technical decision: whether to fund UHTREX or the Clinch River Breeder reactor. The latter won, but was itself was cancelled several years later.
By resurrecting this thinking, which is still valid, it could lead to a new reactor type, one where capital costs should be quite low, as the core dimensions might be about the size of a 55-gallon barrel instead of 20 to 30-feet for a large-scale nuclear power plant. This small size should make it easy to secure in a fixed site, a definite plus with worries over terrorism. The high temperature gives greater efficiency in producing electricity while the small size means it can even be put on ships, barges or even trucks and moved around the country in emergencies, say if a tornado, hurricane or flood knocks out power to a grid. It might be useful for unexpected summer brownouts or blackouts also. Once again, this is not anything original on my part, but only looks at what had been done successfully. See the USS Sturgis. The high temperature also opens up its possible use in chemical, petrochemical, steel and other industries that require large amounts of process heat. That could be a large market. Most of all, it could be decidedly green by opening up a third way of handling the wastes that are clogging our landfills or polluting our oceans: with its blast furnace temperatures, just melt the wastes and then separate the liquid into the commercially valuable and the true waste. The former is re-used, the latter disposed of, perhaps permanently in space. If this is feasible Ė I think it is, but its never been studied to my knowledge - I believe the greens would welcome this Ė it makes sustainable development a reality - while politicians would clamor for it - itís their dream come true - as it brings jobs and industry to Rust Belt districts. This could be an extraordinarily large and profitable market. I discuss this more in The Nuclear Rocket.
In conclusion, Iíve covered a lot of ground and hope Iíve persuaded all to call for an end of the ban on using nuclear rockets to reach and return from LEO. By now all should have nuclear rockets in their bones and see how dynamic they are. Itís a game-changer for space. Look at your payload fractions and compare with chemical ones. Though conceptual now, they promise to change the space program from an elitist, government/corporate effort to one for the common man, one open to you, for you to send payloads there first and later for you to go there and not for $50 million per ticket. And you can own, run and profit from it, not only in space but also on earth. So organize, form study groups, read the full argument in The Nuclear Rocket. Start debate and dialogue, thereís plenty to do. You are the future, once you demand a new way of thinking, the system will begin to change. Then ask NASA Ė in letters, in questions in public meetings, in conferences Ė to state its reasons for not considering the use of nuclear rockets to reach and return from LEO. Without that pressure, without groups agitating in the public, without some in the public sector giving them political cover to change, the space program bureaucrats wonít do anything except what they have been doing. As a former bureaucrat myself, I can tell you the prevailing adage isĒ the exposed nail gets hammered down.Ē They will keep hunkered down and never look at using nuclear rockets to reach and return from LEO unless there was a loud and vocal public interest in it. Theyíll keep plugging away at manned Mars like they have for the last half a century and keep getting the same result: NO!
And safety? Well, as you work through each phase of launching and recovering a nuclear engine you will see ways to make the risks quite manageable, like those we take in flying from NY to LA. Even those most likely to oppose initially will, I believe, come to be its strongest supporters, as they see it will allow them to accomplish their goals of greening the planet and making it peaceful and prosperous. Itís a game-changer for earth as well.