The Case for Space IX: Fission Fragment Rocket

Fission Fragment Rocket Engine (FFRE)

1. Description
2. Design
3. Advantages
4. Disadvantages
5. Economics
6. Summary


If you would enjoy more details on this concept, go to the Atomic Rockets.  I keep this website open in a tab in my browser for quick reference and enjoy randomly selecting articles and reading them even if it isn’t relevant to my current research.  Even though I had heard about Fission Fragment Rockets before finding Atomic Rockets page, much of my current information on Fission Fragment Rockets come from these pages.

Details about a mission specific FFR can be found at the Next Big Future: Dusty Plasma Based Fission Fragment.


1. Description

In most nuclear engine designs, the nuclear reactor generates power which heats a working fluid / propellant for use in the engine.  However, the fission reaction products (fission fragments) move at a significant fraction of the speed of light (3% to 5% of c).  If instead of using these fragments to heat a propellant, what if we only used fission fragments as propellant?  We’d get a very high specific impulse engine that generated very little thrust.  You’d need an extremely high power fission reactor to generate enough fragments for the necessary thrust.

The Fission Fragment rocket differs significantly from other concepts in a number of ways.

• Spectacular fuel efficiency – any location within the solar system consumes less than 1% spacecraft mass for fuel
• Minuscule thrust - amounting to 2 ten thousandths of a gravity – changing the way we would use the engine.
• Massive secondary engine components – notionally 90% of the spacecraft mass

Back to the top

2. Design

The original concept used a fuel consisting of a small grained dust of Uranium 233 or 235 or Plutonium 239.  Surround this fissile dust with powerful magnetic fields generated by superconducting magnetic coils. Since neutrons are not affected by magnetic fields, line the cylindrical chamber’s physical surface with a light weight neutron reflective material to slow the neutrons and reflect them back into the engine chamber to stimulate more fission reactions.  Close one end of the chamber using a magnetic mirror (may also be used as a direct power generator).  Leave the other end of the chamber open to space – this is your rocket exhaust.

Dusty plasma bed reactor

A fission fragments ejected for propulsion
B reactor
C fission fragments decelerated for power generation
d moderator (BeO or LiH)
e containment field generator
f RF induction coil

The Reaction

In this engine, when a U235 nucleus gets hit by a neutron, it decays into fast moving lighter element (such as Krypton and Barium), 3 more neutrons, and some gamma radiation.  With a typical FFRE reaction looks something like:

n + 235U -> 3n + 138Ba + 97Kr + 1-3 gamma

Note that the actual products, their isotopes, and the number of gamma rays generated can vary between reactions of the same parent atom.

Nuclei Products

These products provide most of our thrust.

The dust size allows the reaction products to escape its surface (and carry away excess heat).  The daughter nuclei possess a charge so when the encounter the containment magnetic field they get turned along the axis of the cylinder.  Approximately ½ of the nuclei turn away from the exhaust of the rocket.  These encounter the magnetic mirror which reflects them back down towards the rocket exhaust.  It is these fragments that provide the rocket with its thrust.  The high velocity of the fragments provides the engine with its high specific impulse.

Calculations of product nuclei thrust suggest an efficiency of 46% (meaning 46% of fragment momentum goes into engine thrust, the rest must be disposed of as waste heat).  The escaping products also carry away significant amounts of heat, reducing the need to expel waste heat by a significant fraction.

Neutron Products

The dust size permits neutrons released by a fission reaction to also escape their dust grain.  Magnetic fields do not influence the motion of the neutrons, so these fly until they hit one of three things 1) the containment vessel wall (generating heat) 2) another dust grain (generating heat) 3) exit through the exhaust (see below).

A small fraction (perhaps 3%) of the neutrons produced can contribute a tiny amount (possibly as much as 1 Newton) to our thrust.  The neutrons hitting other dust grains (perhaps 48%) either increase the fission rate of other Uranium atoms or generate heat.  The remaining neutrons (49%?) hit the beryllium wall of the engine chamber.  As the power of the engine increases, more neutrons hit dust grains and fewer hit the engine chamber.

Because this engine generates so many neutrons, designers must include special provisions for shielding the payload (humans?).  I’m working on another post discussion radiation and shielding.

Gamma Rays

In addition to the daughter products and neutrons, fission reactions generate gamma rays – which are also difficult to stop.  As with the neutrons, the magnetic containment field doesn’t deflect gamma rays so about 3% of these escape through the exhaust and provide a small boost to the propulsion.  All remaining gamma rays impact the containment vessel wall shielding (generating heat) or other fuel dust particles.  These products primarily generate heat in the chamber wall or the dust particles.
A small fraction (3% of them) escapes and contributes a minuscule amount to our thrust (about 1/3 of a Newton).

Like neutrons, gamma rays are highly penetrating and require special shielding to protect sensitive portions of the ship (like the human crew).  I’ll write more about radiation shielding for gamma rays in a later post.

Fuels

The three best fuels for this concept are

• Uranium 233
• Uranium 235
• Plutonium 239

Of these fissionable materials, Plutonium possesses the most energetic fission while Uranium 233 possesses the least energetic (about 2% difference in energy).

However, any radioactive material which primarily generates charged particle products (nucleus fission) or helium nuclei (alpha decay) during nuclear decay could be used for this purpose.  I haven’t calculated the energy available in non-fission spontaneous decay reactions (nor have I found any references which have).  I suspect that alpha decay particle emissions would provide dramatically inferior thrust to the fission product reactions.

Uranium 233

Uranium 233 produces slightly inferior thrust (by about 2%), however, there is much more of it available (via breeder reaction from Thorium 232).

Uranium 235

Uranium 235 produces average thrust but is the least abundant.  It also requires a great deal of work to separate it from the more abundant Uranium 238 isotope.  A potential advantage of using Uranium 235 mixed with Uranium 238 is that the much of the Uranium 238 will transmute into Plutonium 239 during the operation of the engine.

Plutonium 239

Plutonium 239 produces the best thrust and is moderately abundant (via breeder reaction from Uranium 238).


Back to the top

3. Advantages

The biggest advantage of this engine concept is the exceedingly high exhaust velocity.  The high exhaust velocity enables craft built around this engine to thrust continuously for any mission within the solar system.

As a comparison, look at the mass fractions required for a conventional chemical rocket versus a FFRE generating the same change in delta velocity.
  

10 km/s	Saturn V	FFRE
Fuel	90%	0.01%
Structure	6%	90%
Payload	4%	10%

The fraction of the FFRE ship’s mass devoted to fuel is so small, that it can reach nearly any place in the solar system without appreciably changing the total weight of the craft.

Type	Mission	Ship mass	Mission Payload	Mission Payload Fraction	Propellant Velocity	Specific Impulse	Thrust	Max delta V	Mission Time
		tonnes	tonnes	%	km/sec	sec	megaN	km/sec	Months
Light Craft Launch		1	0.9	90%	N/A			10	<1 span="">
Light Gas Gun Launch		0.5	0.15	30%	N/A		1.1	7/10	<1 span="">
Ram Accelerator Launch		2	0.8	40%	Variable		.058 - 250	8/10	<1 span="">
Nuclear Pulse Propulsion	Minimum Launch	300	51	17%	30	3,061	8	12	<1 span="">
	Midsize Launch	2,000	1,080	54%	60	6,122	53	24	<1 span="">
	Inter planetary Launch	4,000	1,600	40%	40	4,082	80	30	<1 span="">
	Advanced Interplanetary Launch	10,000	6,100	61%	120	12,245	400	100	<1 span="">
	Max Launch mass	8,000,000	4,320,000	54%	60	6,122	213,514	24	<1 span="">
	Max Impulse to Solar Focus	10,000	2,135	21%	9800	1,000,000	8	6343	41
NTR	Launch	603	150	25%	9.5	969	0.33	3	<1 span="">
	LEO to Mars	700	150	21%	9.5	969	0.33	3	5
	Launch to Mars one way	15,555	150	1%	9.5	969	0.33	3	5
Fission Fragment	Solar Focus (550 AU)	21,350	2,135	10%	14,700	1,500,000	0.00002	6000	120
	Oort Cloud (0.5 LY)	79,074	2,135	2.7%	14,700	1,500,000	0.00034	18000	360
Chemical Launch	Saturn V	3,350	131	4%	2	263	34	10	<1 span="">
	Space Shuttle	2,030	27	1%	3.5	455	31	10	<1 span="">
	Shuttle C	2,030	100	5%	3.5	455	31	10	<1 span="">

NOTE: The solar focus is an interesting astronomical phenomenon. Gravity bends light and the Sun's gravity is no exception. If you can travel to the correct distance from the sun (550-800 astronomical units) and look back, you can use the sun like a giant telescope lens. With this lens you can get super high resolution images of objects on the other side of the Sun. This mission would have the side benefit of providing a very long baseline for interferometer measurements as well as providing excellent precision parallax (distance to the stars) measurements. It would take about 4 years of travel time (2 years to go out and 2 years to get back) to complete this mission with a nuclear pulse propulsion or 10 years to do it with a Fission Fragment Rocket. 

Most equipment reused

For most interplanetary missions, this craft consists of a large fraction (67-90%) of structures, a smaller fraction of payload (10 – 30%), and a tiny fraction (remainder) of fuel.

Because so much of the structure is reusable (engine, magnets, neutron reflector, heat radiators, etc.), the entire craft should easily be reusable after refueling.

Must carry its own propellant

For interplanetary missions, the mass of the fuel required is so small that it’s less than a rounding error on the mass of the other components.  Because the fuel mass fraction is so small, having to carry it on the missions doesn’t impose the tremendous burden that chemical or other propulsion methods do.

Can use Cheap Materials

Because this engine doesn’t require very high mass fractions, the design can make do with lower performance parts than those required for current generation chemical rockets.  Lithium (coolant), beryllium (neutron reflector), water (neutron & gamma ray shielding), and super conducting magnets (magnetic nozzle) compose the bulk of the FFR engine.  These and the ancillary equipment that go with it consist of between 65 and 90% of the total mass of the craft and all of these are reusable and fairly cheap as far as aerospace materials.

Requires no technological breakthroughs

Although the original concept this rocket developed in the 1980s, it doesn’t include anything that requires more basic research.  Humans have already demonstrated the individual components required for to make this engine (heat radiators, nuclear reactors, breeder reactors, neutron reflectors, liquid lithium coolant, etc.).  We have not tested them in a combination which makes an engine.
However, unlike all other concepts I’ve written about no test articles have been created to test the elements of the design.  We can assume that most elements of the design require more development and testing before moving into full scale design and production.

Large payloads

This engine isn’t able to perform launch operations, there’s really no limit to the mass that the engine can drive.  However, moving large masses will require the production of large quantities of fuel to power the engine.

Interplanetary flight

This concept can’t launch mass off the Earth and there’s little benefit to using this engine in the Earth – Moon system.  Therefore, this engine would be used primarily for interplanetary missions.  It gives superior performance (reduced transit costs and reduced transit times) to almost any other design.  Only the Nuclear Pulse Propulsion (aka Project Orion) type propulsion outperforms this design.

Estimated cost per pound

See the section on 5. Economics.

Back to the top

4. Disadvantages

There are a number of unusual disadvantages with this design.  The two biggest disadvantages turn out to be its exceedingly low thrust and the enormous mass of the engine required to generate modest thrusts.

Rapid turn-around time

The availability of fuel and refueling time, drive the craft’s turnaround time.  Special docking and berthing considerations must be taken to ensure the refueling facility / space station performing the operation is not irradiated by the engines.  Operators must use special handling equipment and procedures to minimize exposure to the engine radiation.

The engine can be shut down when not needed for propulsion or power.  This will substantially reduce the amount of radiation emitted by the engine but neutron activation of some parts will occur and we can expect the engine to remain “hot” for some time after engine shut down.  Since light elements compose most of the engine parts, we can expect the amount of radiation to drop quickly over time.

So little fuel is consumed during most voyages, one option to reduce the turn-around time of this design would be to fuel at the end of a round trip voyage or perhaps refuel after two or more round trip voyages.

Despite the poor turn-around time, this design would possess a very high operation percentage (operating time / total time) due to the amount of time it would spend on missions.

Not Safe for nearby spacecraft

In order to get the minuscule accelerations mentioned in this entry, this ship requires a truly massive nuclear reactor.  The minimum size reactor for 1 tonne of payload started at about 1 GW.  The reactor required for more massive payloads would range substantially higher.

Because it’s very expensive to shield such powerful reactors in all directions (that eats directly into available payload mass), the ship would utilize a shadow shield to protect the vulnerable portions (and people) of the spacecraft.

This means that anywhere not protected by the shadow shield (i.e. in front of the ship), would be bathed in neutron particle and gamma ray radiation.  All docking and rendezvous maneuvers would have to keep the other craft in front of this FFRE craft to ensure the reactor didn’t kill the other crew.

This ship is NOT safe for nearby vessels.

Gentle acceleration

A 9 tonne engine of this design generates a meager 22 Newtons (5 lbf) of thrust.  If the total mass of the craft were 10 tonnes, then the acceleration generated would be about 2/10,000 g.
Since the delta V requirements listed for various destinations count on minimum energy (for a direct transfer) Hohmann transfer trajectories, these trajectories require velocity changes impulses applied at specific points, and it takes this engine months to generate that amount of delta V; then the total energy required for these velocity changes would be measurably larger.

This design also has a benefit to offset this disadvantage; primarily that the engine can operate continuously while performing orbital transfers.  For missions outside the Earth-Moon system this reduces the transit time tremendously, despite the 19 days it takes this vehicle to achieve Earth escape velocity.

			Hohmann Transfer		Achieve	Constant
	dV	Distance	Time		delta V	Acceleration	FFRE mission time
Destination	km/sec	km	days	months	days	days	days	months
Earth Escape	3.6				19	-	19	0.6
LEO	10.0				53	-	53	1.8
GEO	3.8	36,129			20	2	20	0.7
L4/L5	4.1	387,097	9	0.30	22	7	22	0.7
Moon	4.8	387,097	9	0.30	25	7	25	0.8
Mars	5.7	48,387,097	510	17	30	77	96	3.2
Venus	6.9	48,387,097	288	10	36	77	96	3.2
Mercury	13.0	96,774,194	240	8	68	109	127	4.2
Vesta	8.7	410,000,000	930	31	46	223	242	8.1
Ceres	9.5	410,000,000	930	31	50	223	242	8.1
Ganymede	21.0	750,000,000	840	28	110	302	321	10.7
Jupiter	24.2	750,000,000	840	28	127	302	321	10.7
Saturn	18.2	1,400,000,000	4,350	145	96	413	432	14.4
Uranus	14.7	2,850,000,000	5,517	184	77	589	608	20.3
Titan	15.0	4,500,000,000	4,350	145	79	740	759	25.3
Neptune	15.5	4,500,000,000	10,686	356	82	740	759	25.3
Pluto	11.4	6,500,000,000	15,873	529	60	890	909	30.3

If you don’t enjoy pouring over tables of numbers, what this shows is that for any mission beyond the Earth – Moon system the FFRE system completes the mission faster than conventional propulsion systems.  The constant acceleration of the FFRE concept provides more benefit for more distant missions.

Despite the 19 days it takes for the FFRE engine to build up enough velocity to spiral out of Earth orbit, the FFRE engine completes even the Mars mission in less than 1/6th of the time it takes a conventional engine.  For longer missions, the disparity between the time it takes to perform these missions is even greater (it takes the FFRE 1/17th of the time).

The gentle acceleration provides mixed advantages and disadvantages.

Launch Vehicle

This engine not only can’t lift a spacecraft from the Earth, it’s unlikely to be able to lift a spacecraft from most substantial bodies in the solar system.  For instance, Ceres’ surface gravity is 100 times the acceleration provided by this engine and even Phobos’ surface gravity is about 3 times greater than the acceleration this engine can provide.  This engine will only ever be used for interplanetary missions and not launch or missions.

You could envision this engine as the tractor trailer hauler craft.  Not glamorous or showy but it would have a tremendous total delta velocity capability.




Back to the top

5. Economics

This engine cannot launch cargo into orbit which means that I can’t compare its launch costs the other propulsion concepts.  In fact the thrust is so low, that it’ll cause the trip times to increase dramatically from the normal Hohman transfers trajectories.  However, the specific impulse is so high, it’ll be able to maintain thrust throughout the entire trip.

Also this engine requires fissile materials.  The cost of Uranium 235 and Plutonium 239 run about $5,600,000 / kg while Uranium 233 costs only about $2-4 million / kg.  To fly a 1 ton payload out to Pluto and back with continuous engine operation the whole way, would consume about 1.2 tons of fuel and cost about $5billion in fuel.  To do the same mission for a 2135 ton payload (some the size of an initial parts of a permanent human settlement), would consume about 2,700 tons of fuel and cost $10 trillion in fuel.

In an attempt to compare apples to apples, it would cost a FFRE 40 kg of nuclear fuel for a cost of $146,000 per ton to generate 10 km/sec delta velocity (about the velocity change required of launch vehicles).  This equals about $146 / kg or $66 / lb to operate.


Back to the top

6. Summary

I've updated this chart so that launch systems with low costs get an extra point.

Positive attributes of craft	LC	LGG	RA	NPP	NR	FFR	STS
Most equipment reused?	Yes	Yes	Yes	Yes	Yes	Yes	No
Doesn't carry its propellant?	Yes	Yes	Yes	Yes1	No	Yes1	No
Rapid turnaround?	Yes	No	No	Yes	No	Yes4	No
Safe for nearby craft?	No	Yes	Yes	No	Yes	No	Yes
Can use cheap materials?	Yes	Yes	Yes	Yes	No	No	No
Requires no breakthroughs?	Yes	Yes	No	Yes	Yes	No	Yes
Gentle acceration3	Yes	No	Yes	Yes	Yes	Yes	Yes
Large payloads (>50 tons)	No	No	No	Yes	Yes	Yes	Yes
Interplanetary flight	No	No	No	Yes	Yes	Yes	Yes
Space launch	Yes	Yes	Yes	Yes	No	No	Yes
Estimated 10km/sec dV cost per pound	<$500	<$500	$500	<$5002	TBD	$66	>$10,000
Score	10	7	7	10	7	7	6

• LC – Lightcraft
• LGG – Light Gas Gun
• RA – Ram Accelerator
• NPP – Nuclear Pulse Propulsion
• NR – Nuclear Thermal Rocket
• FFR – Fission Fragment Rocket
• STS – Space Transportation System aka Space Shuttle

1. Nuclear pulse propulsion has such good performance that this really isn't a drawback for that concept.
2. I'm making a WAG (wild @$$ed guess) at this because I have no idea how much nuclear bombs cost.
3. This is a measure of whether the system could be used for manned vehicles. Many of these systems can be tuned to provide different accelerations.
4. Infrequent refueling, however, that refueling would take some time.

Back to the top


Proceed to my next space blog, (not completed yet).

Return to my previous space blog, The Case for Space VIII: Nuclear Pulse Propulsion.