External Pulsed Plasma Propulsion (EPPP) is a space propulsion concept that gets thrust from plasma waves generated from a series of small, supercritical fission/fusion pulses behind an object in space.
This NASA study from 2000 had the following designs:
The realistic maximum Isp obtainable with fission-based EPPP is -100,000 seconds. [A fusion powered version of Orion would max out at 1,000,000 ISP (this link compares Orion versions to other space propulsion including photonic propulsion)] However, this type of performance would only be possible with very large spacecraft. Such vehicles would be impractical until the cost of access to space dropped substantially or in-space manufacturing became available. Therefore, a more conservative approach has been taken by considering smaller vehicles with lower performance (Isp 10,000 seconds) using technology available in the near-term. This concept has been informally termed “GABRIEL.” The GABRIEL series includes an evolutionary progression of vehicle concepts that build upon the nearest-term implementation of EPPP. This concept roadmap eventually culminates in larger systems that employ more sophisticated methods for pulse initiation and momentum transfer. GABRIEL is characterized by the following four levels:
1. Mark I: Solid pusher plate and conventional shock absorbers (small size)
2. Mark II: Electromagnetic coupling incorporated into the plate and shocks (medium size)
3. Mark III: Pusher plate extensions such as canopy, segments, cables (large size)
4. Mark IV: External pulse unit driver such as laser, antimatter, etc. (large size)
All of these levels, besides the GABRIEL Mark I, require technology that is not currently available, but may be attainable for a second-generation vehicle. The Mark I is also the smallest and least expensive version, but suffers from the poorest performance (nominally 5,000 seconds and 4 million newtons of thrust).
Note: this poorest performance is still over ten times better than chemical rockets and is five times better than nuclear thermal solid core. The ISP is the level of good ion drive but has thrust that is over 10,000 times better.
The classic version of this idea is Project Orion, which was the idea of firing about 800 small nuclear bombs through a hole in a large metal plate.
An imagining of Project Orion [wikipedia from this link] taking off
Sam Dinkins notes:
One of the 1958 designs achieves an Isp of 12,000 seconds with an Earth-launched payload capacity to LEO of 5500 tons. The fuel would be 800 nuclear bombs. The weight of 800 nuclear bombs has gone way down. For the eight terajoule (two kiloton) explosions required for the 4500-ton version that can carry 20 people to the Moon, Jupiter, and Saturn in the same trip, instead of the 900 tons of nuclear bombs at almost one ton each, there would need to be probably only 50 tons’ worth of bombs [because of technological improvements]
To protect the people from the huge kick of a blast, Orion envisioned huge shock absorbers designed to absorb the impulse of the 1000-ton pusher plate with the entire rest of the ship. Other versions looked at further absorption just in the crew area. My recommendation would be to put the crew on a long electromagnetic track. They could potentially be the only part of the ship that is isolated from the high-g shocks. By having a very small mass isolated from the pusher shocks, the mass of the shock absorbers could be reduced from 900 tons to something more manageable.
Apollo size rocket on the left and a first smaller version of Orion nuclear rocket
Metals have been strengthened with nanograin structure that are four times stronger while not becoming brittle
Here is an article that examined the mass fraction and payload that different versions of Project Orion could launch.
Space bombardment blog also examines the details of the 880ton 1959 design Improved pusher plate and lighter bombs would have it launch 650 tons to orbit instead of 300 tons.
Pictures of Project Orion
The project Orion page
Rapid fire z-pinch fusion is being developed
The Z-pinch is the basis for the minimag Orion concept
Mag Orion would have detonated 100 kiloton bombs 2 kilometers behind the space craft and had a superconducting magnetic sail interact with the blast to generate 1,000,000 newtons at 30,000 ISP.
Mini-mag Orion would use sub-critical explosions with Z pinch technology. 5 tons of explosive power with a 5 meter magnetic bottle. Specific Impulse of 21,500 sec and thrust 625,000 Newtons.
A review of NERVA nuclear rockets and variants of NERVA
Nuclear thermal rocket like NERVA
Most studies of nuclear thermal (NERVA variants) believe that an ISP of 925 would be very achievable. Russian NTR fuel elements would allow ISP of 960+. A 2005 NASA presentation on nuclear thermal Systems with ISPs of 1010 considered. There are designs of closed cycle gas core versions with 1500-2000 seconds (ISP) (15–20 kN·s/kg). Gas core reactor rockets can achieve 3000 to 5000 seconds (ISP) (30 to 50 kN·s/kg).
The disadvantage of the open cycle is that the fuel can escape with the working fluid through the nozzle before it reaches significant burn-up levels. Thus, finding a way to limit the loss of fuel is required for open-cycle designs. Unless an outside force is relied upon (i.e. magnetic forces, rocket acceleration), the only way to limit fuel-propellant mixing, is through flow hydrodynamics. Another problem is that the radioactive efflux from the nozzle makes the design totally unsuitable for operation within Earth's atmosphere. The advantage of the open cycle design is that it can attain much higher operating temperatures than the closed cycle design, and does not require the exotic materials needed for a suitable closed cycle design.
So if you are going for high performance you might as well go for EPPP instead of open cycle gas core nuclear thermal.
Lunar resource overview
Lunar resources could be developed so that EPPP rockets could be built there and launched to the rest of the solar system. The next step is to build upon a plan that I have proposed for taking space based solar power profitably to the megawatt level Then developing a lunar base with hundreds of megawatts to gigawatts of lightweight solar power. Some solar cells and reflective material could be built using lunar regolith. After the industrial and energy base is established then the mining, industry, nuclear power plants and nuclear rockets could be built.
Mirrors and lasers for photonic propulsion should also be constructed
Studies of the lunar regolith indicates a lot of Thorium, although the exact amount is uncertain
There is Thorium in the lunar regolith which can be put into a breeder reactor to make Uranium 233.
The Uranium 233 can make bombs but it would more difficult than Uranium 235 because of the radiation.
Thorium can be used to power nuclear reactors that are superior to current nuclear reactors
Here is an analysis of lunar regolith
Composition of lunar regolith
KREEP ore deposits on the moon would include Chlorine, Zirconium, Fluorine, Thorium, Potassium, Uranium, Phosphorus, Rare Earth Elements, Sodium
Thorium as a tracer for KREEP deposits on the Moon. Note the high concentrations around the Procellarum KREEP Terrane.