Direct kinetic-to-electric conversion through magnetohydrodynamic (MHD) fission fragment streaming
© 2026 David R. Young — Spectrum Energy Research Foundation · CC BY-NC-SA 4.0
© 2026 David R. Young — Spectrum Energy Research Foundation · CC BY-NC-SA 4.0
Every semiconductor — selenium, silicon, diamond, cadmium telluride — converts kinetic energy to electron-hole pairs at the same fundamental ratio: three electronvolts spent for every one recovered. The other two become lattice heat. That is physics, not a material limitation, and it caps electrical conversion at 33% regardless of the semiconductor chosen. But fission fragments are not neutral particles. They emerge carrying twenty to forty positive charges, moving at three percent of the speed of light. What happens if you stop treating them as particles to be absorbed and start treating them as a current to be harvested?
When a fission fragment tears through a semiconductor, it rips electrons away from atoms along its path. Each separation creates one electron-hole pair — one unit of recoverable charge. But creating that pair costs about three times the energy the pair carries. The excess goes into rattling the lattice — phonons, vibrations, heat. This ratio is built into how charged particles interact with crystal lattices. It holds in selenium, silicon, diamond, and every other semiconductor material. No engineering, no geometry, no electrode placement changes it. The ceiling is 33%.
With perfect carrier collection — every pair reaching an electrode, zero recombination — a semiconductor converts 33% of fragment kinetic energy to electricity. The other 67% is heat. The Warm Fission Reactor's working assumption of 25% means real-world recombination losses subtract another 8 percentage points below the theoretical limit. Closer electrode spacing can recover some of that gap, but can never reach above 33%.
This ceiling applies to 82.5% of all fission energy — the fragment kinetic channel. It is the dominant energy band. Any path to higher electrical efficiency must leave semiconductors behind for this channel.
When a uranium-235 atom splits, the 92 protons divide between the two fragments — roughly 38 in the light fragment and 54 in the heavy fragment. But the 92 electrons do not follow neatly. The violence of the split strips most of them away. Each fragment emerges carrying a net positive charge of +20 to +40 — twenty to forty electrons missing.
This charge is not something that needs to be created. It is intrinsic to fission. Every fission event produces two massive, fast, highly charged positive particles and a cloud of free electrons — already separated, already in motion.
A particle with a charge moving at high velocity is a current. Two fragments per fission, billions of fissions per second — that is not a trickle. It is a dense, fast, positive current, delivered free by the reaction.
A charged particle moving through a magnetic field curves. The stronger the field, the tighter the curve. The heavier the particle, the wider the curve. The exact radius depends on three things: the particle's mass, its velocity, and its charge.
A light fission fragment at 100 million electronvolts with a charge of +20, in a 5-Tesla magnetic field, curves with a radius of about 14 centimeters. This is a manageable dimension — smaller than a dinner plate.
Now picture a cylinder with uranium fuel on its inner wall. A magnetic field runs along the cylinder's axis, generated by a solenoid coil wrapped around the outside. A fission event occurs at the wall. The fragment launches inward, carrying its charge at enormous speed. The axial field catches it and curves it into an orbit inside the cylinder. The fragment circles, passing the cylinder wall repeatedly.
Each time the orbiting fragment approaches a wall electrode, it pushes electrons through the external circuit. As it curves away, the current reverses. Each orbit pumps charge back and forth — alternating current at the fragment's natural orbit rate. The fragment loses kinetic energy to the circuit with every pass. The orbit shrinks as the fragment slows. Eventually it stops and gets collected.
No channel. No axial flow. The cylinder wall IS the generator. The orbiting fragment is the moving charge. The wall electrodes are the pickup. This is Faraday's law of electromagnetic induction applied to a circling ion instead of a spinning rotor — the same principle behind every electrical generator ever built.
The extraction mechanism is the reverse of a cyclotron accelerator. A cyclotron kicks a particle faster each orbit with a timed electric pulse. This design brakes the particle each orbit by extracting current from its motion. Cyclotrons have worked since 1932. The physics runs both ways.
The dual-uranium membrane emits fragments both directions. Inward fragments orbit inside the fuel cylinder. But outward fragments fly through the diamond and exit the cylinder into the space beyond.
The magnetic field does not stop at the cylinder wall. It fills the entire solenoid volume. The outward fragment is just as charged as the inward one. The moment it exits the membrane, the field catches it and curves it into an orbit outside the fuel cylinder.
A second, larger cylinder surrounds the fuel cylinder, with electrodes on its inner face. Outward fragments orbit in the annular space between the two cylinders. Same physics, same cyclotron frequency, same extraction mechanism. Just a larger orbit radius.
The geometry is three concentric zones:
Inner zone: Open space inside the fuel cylinder. Inward fragments orbit here. Pickup electrodes on the inner surface (or a central electrode structure) extract energy from the orbiting fragments.
Middle: The fuel cylinder — U-Diamond-U membrane. Fission events emit fragments both inward and outward. Both directions enter vacuum and get captured by the magnetic field.
Outer zone: Annular space between the fuel cylinder and the outer electrode cylinder. Outward fragments orbit here. Pickup electrodes on the inner face of the outer cylinder extract energy.
One hundred percent of fragments are captured — both directions. No 50% rod loss. The fuel cylinder floats in the middle of its own generator, feeding orbiting charges to both sides.
The stripped electrons, carrying negative charge, orbit in the opposite direction from the fragments and at much tighter radius (an electron is 170,000 times lighter). Collector electrodes in the inner and outer zones capture these electrons separately. Two charge polarities, two circuits, one fission event.
The cyclotron frequency for the light fragment at 5 Tesla is approximately 1.6 megahertz. In the non-relativistic regime, this frequency stays constant as the fragment slows — only the radius changes, not the orbit rate. A fixed-frequency extraction circuit works throughout the entire deceleration. This is standard radio-frequency electronics.
Fragment energy
Orbit radius (5 T)
Orbit diameter
100 MeV (full speed)
14 cm
28 cm
50 MeV
10 cm
20 cm
25 MeV
7 cm
14 cm
10 MeV
4.4 cm
~9 cm
0 (stopped)
0
Collection point
Two extraction approaches work with this geometry:
Passive — induced current. Segmented electrodes on the cylinder walls, connected through an external load. The orbiting fragment induces alternating current at its cyclotron frequency. No active electronics needed. The wall electrodes are conductors. The physics does the work.
Active — RF braking. Electrode segments driven at the cyclotron frequency, timed to oppose the fragment's arrival at each segment. Like a cyclotron running backward. More complex but potentially faster extraction — the braking force is tunable.
No electron-hole pairs are created. No lattice absorbs two-thirds of the energy as heat. The kinetic energy of the orbiting fragments transfers directly to electrical current through electromagnetic induction. The efficiency ceiling is set by ohmic losses in the electrodes, field geometry, and the completeness of the deceleration — engineering parameters, not fundamental physics limits.
At the center of the inner zone and at the inner edge of the outer zone, the fragments have zero velocity. The electrodes extracted their kinetic energy before they arrived. They are not hot. They did not impact a wall. They did not thermalize. They are heavy ions, sitting motionless in the magnetic field, at ambient temperature.
And the magnetic field has already sorted them.
Different isotopes have different masses. Different masses orbit at different radii in the same field. Cesium-137 (mass 137) follows a wider orbit than strontium-90 (mass 90), which follows a wider orbit than krypton-85 (mass 85). As the fragments decelerate, each isotope's shrinking orbit converges on a different final position — grouped by mass range through the same physics that captured their energy.
This is exactly how the Manhattan Project separated uranium isotopes at Oak Ridge — electromagnetic separation by mass-to-charge ratio, using devices called calutrons. The same principle, running in the other direction: instead of separating fuel going in, this separates products coming out.
Collection bins positioned at the calculated radii receive pre-sorted, cold, neutral atoms. Electrons are injected at each bin to neutralize the ions. Barium in one bin. Strontium in another. Cesium in another. No chemical processing. No centrifuges. No contaminated liquids to filter. The magnetic field does it for free because it was already there for the MHD conversion.
These atoms are Spectrum Energy Cell fuel. Each isotope can go directly into SE Cell containers — packaged on-site and shipped to customers as a continuous product stream, not a batch process. Short-lived isotopes are immediately active. The reactor's waste stream is the battery's fuel supply, delivered cold, pre-sorted, and ready to use.
The fuel sits on the inner surface of the middle cylinder: enriched uranium deposited on both sides of a diamond membrane, exactly as described in Note 038. The diamond provides the structural base — it holds the uranium atoms in a fixed lattice. Uranium faces both inward and outward, presenting fuel surface to both the inner and outer vacuum zones.
Both zones are vacuum. This is essential — any gas would slow the fragments and thermalize them before the electrodes could extract their energy. The fragment must fly free from the fuel surface into the field with nothing in its path.
Neutrons, which carry no charge, are unaffected by the magnetic field. They fly outward through the thin membrane (nearly transparent at 20–25 micrometers), pass through the outer electrode cylinder, enter the moderator beyond, slow down, and return through the same path to sustain the chain reaction. Zirconium hydride (ZrH₂) moderator rods positioned in the dead zones between fragment orbits — both inside and outside the fuel cylinder — provide additional moderation inside the vacuum volume. These rods are protected from fragment impact by the magnetic field geometry and are unaffected by the field (electrically neutral). Beryllium metal in the external reflector provides neutron multiplication.
A single fuel cylinder may not contain enough uranium to sustain a chain reaction on its own. The membrane is thin — 20 to 25 micrometers — and a thermal neutron passing through it has less than a 1% chance of causing fission. Multiple cylinders solve this the same way conventional reactors use multiple fuel assemblies: add fuel elements in a moderated array until the system reaches criticality.
Coaxial stack: Multiple fuel cylinders arranged along the same axis, sharing the same solenoid. Moderator material — heavy water, beryllium blocks, or ZrH₂ — fills the space between cylinders. Neutrons from one cylinder fly outward, moderate in the gap, and return to the neighboring cylinder. Each cylinder has its own inner and outer extraction zones. One solenoid, multiple fuel sources. Adding cylinders extends the stack until k reaches 1.0.
Parallel stacks: Multiple solenoid-and-cylinder assemblies side by side, sharing a common moderator pool. Neutrons from one stack moderate in the pool and contribute to neighboring stacks. This is how conventional multi-unit reactor stations work — units sharing a pool for neutron economy and cooling.
Power scaling: Either arrangement gives a power scaling knob. More cylinders, more orbiting fragments, more electrical output. Capacity grows by adding fuel cylinders to the stack or parallel units to the pool — without redesigning the extraction electrodes or the coil.
A short, wide cylinder — a disk — may be the optimal individual fuel element. Closer to a sphere's low surface-to-volume ratio (which minimizes neutron leakage), but still cylindrical for the axial field to work. The moderator and reflector surrounding each disk compensate for whatever leakage the geometry does not prevent. The exact dimensions — cylinder diameter, length, spacing, and number — are a neutronics optimization for MCNP modeling.
Each energy band couples at its own layer. Nothing needs to escape the shielded volume. The fragment stream, the neutron cycle, and the gamma conversion all operate at different radii without interfering with each other.
Layer 1 — Inner vacuum zone. Inside the fuel cylinder. Inward fragments orbit here, decelerating against pickup electrodes. Electron collectors capture stripped electrons orbiting in the opposite direction. ZrH₂ moderator rods in the dead zones between fragment orbits. Fragment collection bins at the center where decelerating orbits converge.
Layer 2 — Fuel cylinder. U-Diamond-U membrane. Fission events emit fragments both inward and outward. Both enter vacuum zones where the magnetic field captures them into orbits.
Layer 3 — Outer vacuum zone. Annular space between the fuel cylinder and the outer electrode cylinder. Outward fragments orbit here, decelerating against pickup electrodes on the inner face of the outer cylinder. Electron collectors and ZrH₂ moderator rods in the dead zones. Fragment collection bins at the outer edge where decelerating orbits converge.
Layer 4 — Outer electrode cylinder. Inner face carries the pickup electrodes for the outer fragment orbits. Structural shell of the vacuum vessel.
Layer 5 — Solenoid coil. Wraps around the vacuum vessel. Generates the axial magnetic field. Superconducting coils draw near-zero power in steady state.
Layer 6 — Moderator. Heavy water or beryllium surrounding the coil. External neutron moderation supplementing the internal ZrH₂ rods.
Layer 7 — Reflector. Beryllium shell returning escaping neutrons back toward the fuel. Neutron multiplication via the beryllium (n,2n) reaction above 1.85 million electronvolts.
Layer 8 — Gamma converters. CdTe or CZT semiconductor panels surrounding the reflector. Direct conversion to electricity at approximately 30% efficiency. Scintillator panels alongside for light export via fiber optic. This layer is completely independent of the cyclotron system — a separate conversion pathway for a separate energy band.
Layer 9 — Shielding. Stops whatever gamma the converters do not capture, plus any remaining neutron flux.
Layer 10 — Thermal boundary. Heat exchangers transfer thermal energy from the moderator and shielding to a secondary loop.
Four products exit the shielding through dedicated paths: electricity on wires (from cyclotron extraction electrodes, electron collectors, and CdTe panels), light through fiber optics (from scintillator panels), SE Cell fuel through a sealed transfer port (cold, partially sorted isotopes from the collection bins), and thermal energy through heat exchangers (from the moderator and shielding). Each energy band meets its converter at the correct layer. Each product exits through its own channel. Nothing crosses a boundary it should not.
Every fission event produces two charges moving in opposite directions in the magnetic field:
The positive circuit (primary power): Two heavy fragments, each carrying +20 to +40 charges, enter the inner and outer vacuum zones and decelerate in one arc through the combined magnetic and electrostatic fields. Wall electrodes at high positive potential capture the fragment's kinetic energy as electrical potential. This is the main power output — 82.5% of fission energy.
The negative circuit (return path): Twenty to forty electrons, stripped during the split, orbit in the opposite direction from the fragments at much tighter radius (an electron is 170,000 times lighter than a fission fragment — orbit radius roughly 50 to 100 micrometers versus 14 centimeters).
With the electrostatic field present, the electron collection path changes from the cyclotron-only concept. The electrons cannot fly radially — the magnetic field curves them into tiny orbits. The electric field pulls them outward (toward the positive wall), but the magnetic field prevents radial transit. In crossed electric and magnetic fields, charged particles drift perpendicular to both — tangentially, around the circumference of the cylinder. This is called E×B drift, and it is well-understood physics used in magnetrons, Hall-effect thrusters, and plasma confinement devices.
The drift direction depends on the electric field orientation. In the outer zone, the electric field points radially inward (from the positive outer wall toward the fuel cylinder). In the inner zone, it points radially outward (from the positive central electrode toward the fuel cylinder). The magnetic field is axial in both zones. The cross product flips: inner zone electrons drift one direction tangentially, outer zone electrons drift the opposite direction. The two populations do not mix — they are separated by zone and by drift direction.
Both populations migrate axially — along the cylinder axis toward the ends. The magnetic field does not constrain axial motion. End-cap collector electrodes at both ends of the cylinder capture the electrons. A small axial bias voltage on the end caps can preferentially pull each population toward a designated end — inner zone electrons to one end, outer zone electrons to the other — shortening their dwell time in the interaction volume and reducing recombination risk with decelerating fragments.
The collected electrons route through the external circuit, doing useful work as current, then travel to the fragment collection bins where they neutralize the stopped fragment ions. The same electrons stripped during fission return to complete the atom. Closed circuit from one fission event — both charges doing useful work on the way.
Some recombination between drifting electrons and decelerating fragments will occur — both populations coexist briefly in the same volume. Each recombination event neutralizes one fragment prematurely, losing whatever kinetic energy it still carried. The rate depends on the electron dwell time (reduced by axial biasing) and the fragment deceleration time (microseconds in the single-pass design). The short fragment transit time limits the exposure window. The recombination loss is a calculable percentage to be included in the overall efficiency, not a design-breaking issue.
The two circuits are naturally decoupled by scale. Fragment arcs are centimeters in radius, completing in microseconds. Electron orbits are micrometers in radius at gigahertz frequency, drifting tangentially at the E×B velocity. They occupy different spatial scales and different frequency bands. The magnetic field separates them radially. The electric field sorts them tangentially. The axial bias collects them at the ends. Three fields, two particle species, one closed circuit.
Each orbiting fragment induces alternating current in a wall electrode at its cyclotron frequency — 1.6 megahertz for the light fragment at 5 Tesla. But billions of fragments orbiting simultaneously at random phases average to smooth direct current. The individual AC signals cancel. The natural output is DC.
Conventional power inverters convert DC to AC at 95 to 98% efficiency. This is the baseline approach — simple, external, commercially available today. But there is a second path.
If the solenoid field strength oscillates slightly — a 1 to 2% modulation at any chosen frequency — every orbit in the system breathes in and out together. Field strengthens, all orbits contract, fragments move inward. Field weakens, all orbits expand, fragments move outward. From the wall electrode's perspective, the orbiting charges are closer at field minimum (stronger induced current) and further at field maximum (weaker induced current). The electrode sees a signal that rises and falls at the modulation frequency. That is alternating current — produced directly, at whatever frequency the field is modulated.
The generator analogy is exact. In a conventional generator, the magnetic field rotates past the conductor. Here, the field breathes and the orbiting charges move radially in and out past the electrode. Different motion, same result — a changing distance between charge and conductor producing alternating current.
The electrons do not couple to this modulation. Their orbits are 50 to 100 micrometers in radius. A 1 to 2% modulation shifts them by 0.5 to 2 micrometers — negligible. Their cyclotron frequency is 140 gigahertz, and between one 60-hertz breath of the field, each electron completes over two billion orbits. From the electron's perspective, the field is effectively frozen. The electron circuit remains DC while the fragment circuit produces AC.
This is selective coupling through frequency mismatch — the same principle the Spectrum Energy Charts map across the electromagnetic spectrum. The fragment couples to the modulation because its cyclotron frequency is in the right neighborhood. The electron does not couple because its frequency is 100,000 times higher. Same field, two particles, two different responses. The modulation frequency is the designer's choice: 50 hertz, 60 hertz, 400 hertz for aircraft systems, or any other frequency the application requires.
Working backward from a target of 100 megawatts electrical at 70% extraction efficiency, the required fission rate is approximately 5 × 10^18 fissions per second, producing roughly 10^19 fragments per second. If each fragment orbits thousands of times before its energy is fully extracted, the number of fragments circling inside the cylinder at any given moment accumulates to enormous densities.
Every one of these fragments carries +20 to +40 positive charges. They repel each other. The magnetic field provides confinement force, but there is a maximum density of like-charged particles that a magnetic field can hold — the Brillouin limit. At 5 Tesla, for fission-mass ions, this limit is approximately 7 × 10^8 ions per cubic centimeter.
At reactor power densities, even 100 orbits to extraction produces a circulating population that exceeds the Brillouin limit by a factor of 200. At 10,000 orbits, the excess is 20,000-fold. The fragments repel each other faster than the field can confine them. The beam blows apart.
This does not invalidate the concept of electromagnetic extraction. It means the fragments cannot be stored as orbiting beams at reactor power densities. They must be extracted fast — in one or a few passes, not thousands of orbits. The energy is still there. The charge is still there. The mechanism must capture the energy before the population builds up.
Instead of letting fragments orbit and slowly extracting energy over thousands of passes, apply a strong opposing electric field that decelerates each fragment in a single arc — less than one complete orbit.
The geometry is a cylindrical capacitor. Two concentric conductors with vacuum between them, held at a voltage difference. For the outer zone: the fuel cylinder surface (uranium, conductive) is one electrode. The outer cylinder wall is the other, held at high positive voltage. The electric field fills the annular vacuum gap. A positive fragment exiting the fuel surface immediately begins climbing the voltage hill — decelerating, converting kinetic energy to electrical potential in the field.
The magnetic field curves the fragment's path into an arc while the electric field decelerates it along that arc. The curve means the fragment traverses more distance through the voltage gradient than a straight line — more deceleration for the same gap width. By the time the fragment reaches the outer wall, it has lost most of its kinetic energy to the field. That energy drives current through the external circuit.
This is regenerative braking. The fragment's kinetic energy transfers to electrical potential through the electrostatic field, exactly as a braking electric vehicle transfers kinetic energy back to the battery through the motor's electromagnetic field. The field is the conversion mechanism, not a power drain.
For a fragment with charge +20 at 100 million electronvolts, a 4-million-volt potential across the gap extracts 80 MeV — 80% of the fragment's energy — in one pass. The remaining 20% can be captured by a lower-voltage secondary electrode or accepted as a small thermal loss at the collection point.
The voltage gradient required: at 4 million volts across a 14-centimeter gap, the field is approximately 29 million volts per meter. In high vacuum, electrical breakdown occurs at roughly 50 to 100 million volts per meter with clean electrode surfaces. The operating gradient is well within the safe range.
The space charge problem disappears. Each fragment is born, decelerates in one arc (microseconds), and stops. The number of fragments in flight at any moment is only those currently in transit — a tiny fraction of what the orbital concept accumulated. No stored beam, no population buildup, no Brillouin limit concern.
Competing losses — gas collisions, charge exchange, fragment-fragment collisions — have almost no time to act. The fragment is captured in microseconds. In the orbital concept, these losses had milliseconds to seconds to drain energy. Single-pass extraction wins the race against losses decisively.
The inner zone uses the same principle: a central electrode rod on the axis, held at high positive voltage. Inward fragments decelerate as they curve toward the center. Both zones extract energy simultaneously from fragments emitted in both directions.
A static electric field costs almost nothing to maintain. Once the electrodes are charged to the target voltage, the field holds indefinitely in vacuum. No ongoing power draw. It is a capacitor — a small amount of energy to charge it once, then it sits. The only steady-state loss is leakage current through the vacuum gap, which in high vacuum is essentially zero.
This is counterintuitive because millions of volts sounds expensive. But voltage is not power. Power is voltage times current. With no current flowing through the vacuum insulation, the power to maintain the field is zero.
The fragments supply energy to the field, not the other way around. Each decelerating fragment pushes electrical potential into the circuit. The field is a net energy producer. The high-voltage supply only needs to compensate for tiny leakage — perhaps 10 to 50 watts against 100 megawatts of output.
The superconducting solenoid for the magnetic field draws near-zero steady-state power (zero resistance). The refrigeration system to keep the coil cold draws roughly 50 to 100 kilowatts. Against 100 megawatts of output, that is 0.05 to 0.1%.
Total parasitic power to maintain both fields: approximately 0.1% of output. The conversion mechanism is essentially free to operate. The cost is capital — building the high-voltage system and the superconducting coil. Once built, the fields run indefinitely while collecting energy from every fragment that passes through them.
Note 028 defined a design axis from all-heat on the left (conventional reactor, everything thermalizes) to zero-heat on the right (no material in the fission path, all energy converted directly). Every Spectrum Energy system sits somewhere on this line.
Position
System
Conversion
Electrical efficiency
Far left
Conventional reactor
Steam turbine
~33% (thermal cycle)
Left-center
Conventional + waste harvesting
Thermal + scintillator
~35%
Center
Warm Fission Reactor (Mark I)
Se-S semiconductor (one side)
~12%
Right-center
Warm Fission Reactor (Mark II)
Se-S semiconductor (both sides)
~17%
Right
Cold Fission Reactor
Electrostatic + magnetic (single-pass)
~40–55%
Far right
Zero-heat target
Theoretical limit
—
The Cold Fission Reactor sits further right than any semiconductor-based design can reach. Not because of better materials, but because of a different conversion mechanism — one that does not create electron-hole pairs and therefore does not pay the two-thirds tax that all semiconductors impose.
And the fission products arrive cold. No thermalization occurs anywhere in the conversion path. The kinetic energy of the fragments becomes electrical current in the MHD channel. The kinetic energy of the electrons becomes current in the perpendicular collectors. The only heat in the system comes from gamma absorption in the shielding and from ohmic losses in the electrodes — both small fractions of total output.
Every energy band from a single fission event (200 MeV total), converted through its optimal pathway:
Energy band
MeV
Conversion
Efficiency
Electrical (MeV)
Fragment KE (82.5%)
165
MHD channel
60–80%
99–132
Electron KE
~3
Perpendicular collectors
~50% est.
~1.5
Gamma (7%)
14
CdTe direct
~30%
4.2
Beta (4%)
8
MHD or Se-S
25–60%
2–4.8
Neutron KE (2.5%)
5
Thermal (D₂O)
0% elec
0
Neutrinos (6%)
12
Lost
—
0
Total
200
107–142
Conservative estimate (60% MHD efficiency): 107 MeV electrical from 200 MeV fission — 53% overall electrical efficiency.
Optimistic estimate (80% MHD efficiency): 142 MeV electrical from 200 MeV fission — 71% overall electrical efficiency.
Compared to a conventional reactor's 33% thermal cycle efficiency and the Warm Fission Reactor's 12–17%, this is a different class of machine.
30-unit station comparison (450 MW fission power)
Design
Electrical output
Thermal output
Light output
Conventional (steam turbine)
~150 MWe
~300 MW (waste)
—
Warm Fission Mark I (Se-S, rods)
~108 MWe
~325 MW (product)
—
Warm Fission Mark II (Se-S, membranes)
~150 MWe
~280 MW
~1.4 MW
Cold Fission Reactor (conservative, 60%)
~240 MWe
~190 MW
~1.4 MW
Cold Fission Reactor (optimistic, 80%)
~320 MWe
~110 MW
~1.4 MW
The same 450 MW of fission power that produces 108 MWe in the Warm Fission Mark I produces 240 to 320 MWe through MHD conversion — more than double to triple the electrical output. The thermal output drops correspondingly because more energy is captured as electricity before it becomes heat. The reactor runs cooler because less energy is wasted. The "cold" in Cold Fission is literal — not just the fuel products, but the entire system generates less waste heat per unit of electrical output than any other fission design.
Note: the conventional reactor's thermal output is waste heat that must be rejected. The Warm Fission Reactor's thermal output is a useful product (heating, desalination, industrial process heat). The Cold Fission Reactor's reduced thermal output is still usable product — just less of it, because more of the energy left as electricity instead.
Note 028 — The Heat-to-No-Heat Spectrum. Established the design axis and the principle that heat comes from thermalization, not from fission. The MHD concept is the realization of that principle: remove the material from the fission path, extract energy electromagnetically.
Note 038 — The Diamond Membrane Fuel Cell. Developed the uranium-diamond-uranium membrane that provides the fuel base for this design. The membrane investigation also revealed the 33% semiconductor ceiling that motivated the search for a non-semiconductor conversion mechanism.
EM Propulsion Vehicle Concept — The solenoid coil that generates the axial magnetic field for the MHD channel is the same component type. Shared engineering development path.
Warm Fission Reactor — The moderator (D₂O or beryllium), reflector, control system (B₄C rods), and safety architecture all apply. The MHD channel replaces the Se-S conversion medium but does not replace the nuclear engineering.
Spectrum Energy Cell — The cascade connection. Fragment collection at the end of the MHD channel delivers cold, neutral fission product atoms directly into SE Cell fuel containers. The reactor feeds the cell. The cell's waste is a stable isotope.
Correction 011 — Spin is steerable by external magnets. Nuclear polarization of the fuel by the solenoid field could preferentially align fission axes perpendicular to the cylinder wall, increasing the fraction of fragments that launch cleanly into the channel interior.
External research that supports elements of this design:
MHD power generation — studied since the 1960s for coal and natural gas plants. The U.S., Russia, Japan, and China have built experimental MHD generators. The physics is proven. The difference here is the working fluid: fission fragments instead of combustion plasma.
Superconducting magnets — MRI machines operate at 1.5 to 7 Tesla routinely. Research magnets reach 20+ Tesla. The 5-Tesla field required for this design is well within current engineering. Superconducting coils consume near-zero power in steady state.
Fission fragment research — fragment charge states, mass distributions, and kinetic energies are among the most precisely measured quantities in nuclear physics, with decades of data from facilities including the ILL LOHENGRIN spectrometer.
Calutron electromagnetic separation — Oak Ridge Y-12 plant, Manhattan Project (1943–1945). Separated uranium isotopes by mass-to-charge ratio using magnetic fields — the same principle by which the MHD channel automatically sorts fission product isotopes at the collection point. Proven at industrial scale.
What is the optimal multi-cylinder configuration for criticality? Coaxial stack with shared MHD channel, or parallel stacks in a common moderator pool? How many cylinders, at what diameter and spacing, with what moderator (D₂O, beryllium, or both)? The disk-shaped cylinder (short, wide) minimizes neutron leakage for a given fuel mass. MCNP modeling is needed to determine the minimum number of cylinders for k = 1.0 at 20% enrichment.
What is the fragment stream density at reactor power? Billions of fragments per second flowing through the MHD channel create a dense beam of heavy ions. The mutual electrostatic repulsion of the positive charges (space charge effect) may limit beam density and require channel geometry compensation. How does this affect MHD extraction efficiency?
Bidirectional or unidirectional flow? Fragments exit both faces of the dual-uranium membrane. The field can direct both streams the same way (one channel) or opposite ways (two channels, two generators). Which is more efficient?
What MHD electrode materials survive in a fission fragment beam? The electrodes sit inside the channel, exposed to the fragment stream. They must be conductive, radiation-hard, and compatible with the vacuum and magnetic field environment.
Does the solenoid field polarize the uranium fuel? If the 5-Tesla axial field aligns fission axes preferentially perpendicular to the fuel surface, more fragments launch cleanly into the channel interior. The field is already there for MHD operation — the polarization benefit would come free. But the degree of polarization at operating temperature (not cryogenic) needs investigation.
How does the gamma energy band integrate? Gamma carries 7% of fission energy. It passes through the magnetic field and the vacuum unaffected. CdTe or CZT panels on the exterior of the assembly can convert gamma to electricity directly at ~30% efficiency, as in the Mark II design. This channel is independent of the MHD system.
What is the electron kinetic energy budget? The stripped electrons carry kinetic energy proportional to their share of the fission momentum. The total is small compared to the fragment energy, but the collection efficiency and circuit design for the electron stream have not been characterized.
Can the channel extract beta particles from fission product decay? In battery mode (after shutdown), the collected fission products continue to emit beta particles (electrons). These are also charged particles in a magnetic field. Could a secondary MHD channel or collector capture beta decay energy from the SE Cell fuel while it sits in the collection chamber?
Every generator ever built works by moving a conductor through a magnetic field. This design does the same thing — but the conductor is an orbiting fission fragment, delivered at three percent of the speed of light, pre-charged by the reaction that created it. No turbine, no steam, no semiconductor. The magnetic field does what magnetic fields do. The fragment orbits until its energy is extracted. Cyclotrons have run this physics forward since 1932. This design runs it in reverse.
© 2026 David R. Young — Spectrum Energy Research Foundation
Licensed under CC BY-NC-SA 4.0 for research and education. Commercial use requires a separate license from Spectrum Energy Research Foundation. Contact: secharts@proton.me
© 2026 David R. Young — Spectrum Energy Research Foundation
Licensed under CC BY-NC-SA 4.0 for research and education. Commercial use requires a separate license from Spectrum Energy Research Foundation. Contact: secharts@proton.me