Date of Disclosure: March 10, 2026
Author: Aleksei Shtorm / Solenopsys
License: CERN Open Hardware Licence Strongly Reciprocal v2 (CERN OHL-S v2)
Purpose: Public disclosure to establish priority date and prevent future patent trolling.
A rocket propulsion system where the structural body of the rocket itself serves as the fuel.
The rocket consists of a 3D-printed porous metal matrix (magnesium alloy) which is simultaneously:
- The combustion chamber
- The fuel
- The nozzle array
There are no moving parts, no separate fuel tanks, no turbopumps, no valves.
The entire rocket is consumed during flight, leaving only the payload module.
This is not a reusable launch vehicle competing with Falcon 9. This is a disposable launch unit — the aerospace equivalent of a lighter. Print, load, fire, done.
Aerospace propulsion. Additive manufacturing of rocket systems.
AI-optimized structural design for ballistic and orbital applications.
Target application: dedicated microsatellite launches, 100–200 kg to LEO.
Conventional rockets suffer from:
- High structural dead weight (tanks, pumps, plumbing ~85-90% of mass)
- Complex assembly from thousands of parts
- Long manufacturing and integration cycles
- High cost per launch driven by labor, not materials
Goal: Reduce cost per kg to LEO by an order of magnitude through radical manufacturing simplification. Not by reusability — by eliminating complexity entirely.
The rocket is not built around a conventional engine. The rocket body itself is the engine, the fuel, and the combustion architecture.
It is manufactured as a 3D-printed porous reactive metal matrix — for example a magnesium alloy structure — whose internal geometry contains a very large number of distributed micro-combustion cells, channels, and nozzle-like passages.
The system does not rely on one main chamber and one main nozzle. Instead, thrust is produced by the combined action of thousands or millions of microscopic reactive elements distributed throughout the body.
The key principle is not perfect stability of each individual cell, but macroscopic stability emerging from the statistical behavior of the whole matrix. Local irregularities, local erosion, or local failure of individual cells do not destroy propulsion — total thrust is the averaged result of a very large population of micro-reactive structures.
The structural body is progressively consumed during operation. This consumption is not random burn-through — it is programmed into the geometry itself.
Porosity, channel density, wall thickness, connectivity, and reactive mass distribution vary along the body so that burn progression, thrust development, and mass reduction are encoded directly into the printed structure.
The rocket is not merely designed in CAD. It is computationally synthesized.
Its geometry is generated through iterative simulation and AI-assisted optimization — because the internal architecture is too complex to be manually designed at full fidelity by a human engineer.
The mission profile is therefore physically compiled into the rocket body:
- the structure carries load
- the structure stores reactive mass
- the structure forms distributed combustion zones
- the structure determines how propulsion evolves during flight
Each rocket is printed as a mission-specific object rather than assembled as a standard vehicle with fixed engine hardware. Different orbit = different geometry = reprint.
Key manufacturing innovation:
- Print head is stationary, fixed at one height
- Build platform descends continuously as layers are added
- Enables fabrication of arbitrarily long cylindrical bodies (10–12+ meters) without height-limited build chambers
- Single printer, single material, single process
- No assembly, no post-processing, no human integration steps
Rocket engine performance scales directly with combustion chamber pressure. SpaceX Raptor holds ~300 bar — a world record among production engines, achieved through staged full-flow combustion and turbopumps operating above 700 bar.
CRB reaches equivalent pressure through geometry alone.
A cylindrical cell of 10 mm diameter printed from Mg-Al-Zn alloy (yield strength ~275 MPa) sustains over 800 bar with a 1.5 mm wall — calculated from thin-wall pressure vessel mechanics (P = 2σt/d). Small diameter is not a limitation. It is the pressure advantage.
Internal walls between adjacent cells carry near-zero differential load: both faces are at combustion pressure simultaneously. These walls function as thermal and flow separators, not pressure vessels. They can be made substantially thinner than 1 mm.
The result is not one combustion chamber. It is an array of thousands of micro-chambers operating in parallel — each holding pressure a conventional single chamber of comparable wall thickness could never sustain. Failure of individual cells is statistically irrelevant to total thrust. The engine does not fail — it gradually consumes itself, as designed.
Uncontrolled propagation of the combustion front through a porous reactive metal matrix under 300+ bar LOX pressure would result in detonation rather than controlled thrust. The solution is a sacrificial PTFE (polytetrafluoroethylene) coating applied to the internal pore surfaces — 15–50 microns, infiltrated into the oxide layer of the metal.
The mechanism operates in three phases:
Cold phase. While the matrix is LOX-saturated and cold, PTFE is mechanically pressed into the metal micropores. Direct metal-oxidizer contact is prevented. No ignition is possible.
Transition phase. As the combustion front approaches, radiant heat and hot gas raise local temperature to 400–500°C. PTFE begins thermal decomposition — an endothermic process that briefly delays metal exposure and absorbs energy from the advancing front.
Ignition phase. Once PTFE has volatilized, hot LOX contacts the bare metal surface. The oxide layer, already cracked from thermal stress, fails. Metal ignites instantaneously and combustion propagates to the next cell.
The propagation speed is governed by the PTFE decomposition rate — not by the speed of sound in the metal. Ahead of the front, intact PTFE on cold metal provides no reactive surface for a shock wave to couple with. Detonation cannot run ahead of the thermal front.
A secondary effect: PTFE decomposition releases fluorine, itself a stronger oxidizer than oxygen. The metal-fluorine reaction contributes additional enthalpy to local combustion, marginally increasing specific impulse.
The result is a self-regulating burn front — stable, progressive, and encoded by coating geometry rather than by active control systems.
The porous matrix is saturated with liquid oxidizer (LOX or equivalent) prior to or at the launch site. The specific loading method is not constrained by this disclosure — viable approaches include pressure infiltration, vacuum-assisted flooding, or staged cryogenic filling at the launch stand.
Both open-ended and sealed-base configurations are considered. In the sealed-base variant, ignition may be initiated by a high-energy pulse (explosive or pyrotechnic initiator) that simultaneously ruptures the base seal and initiates combustion across the matrix face.
The absence of fill lines, valves, or pumps in the propulsion path is the key property — not the specific loading procedure.
Unlike conventional solid rocket boosters based on ammonium perchlorate and toxic hypergolic systems, the Al-Mg-Zn + LOX propellant combination is significantly cleaner in terms of combustion products.
The main reaction products are stable metal oxides:
- aluminum oxide (Al₂O₃)
- magnesium oxide (MgO)
- zinc oxide (ZnO)
Thus, the system does not produce the chlorine-containing toxic combustion products characteristic of a number of traditional rocket propulsion schemes.
The booster is a large array of integrated micro-nozzles built directly into the reactive load-bearing structure of the body.
To compensate for changing atmospheric pressure during ascent, the design uses an altitude-gradient distribution of nozzle sections:
- nozzles with lower expansion ratios operate at liftoff in dense atmosphere
- nozzles with higher expansion ratios engage progressively as altitude increases and external pressure drops
- thrust is produced by the combined action of a large number of nozzles with different geometries, pre-distributed throughout the booster structure
Thus, altitude compensation is achieved not by moving mechanical parts, but by the nozzle distribution embedded in the printed matrix.
Three hybrid fin/thruster assemblies mounted on payload module:
- In atmosphere: Aerodynamic control surfaces
- Above atmosphere: Gas-dynamic vanes in exhaust stream
- Fine correction: Micro cold-gas thrusters
- Primary stabilization: Gyroscopic spin from helical surface ribs printed integrally — passive, no actuators
The payload module (satellite bus) is not a separate launch component. It is manufactured by the same LPBF process from the same material family, snaps onto the rocket body, and remains on orbit as a functional spacecraft after the propulsion body is consumed. Its cost is attributed to the satellite, not to the launch vehicle.
The following combinations are claimed as novel prior art:
- Rocket where structural body = fuel (magnesium matrix consumed during flight)
- AI-synthesized fractal nozzle matrix where burn profile encodes trajectory
- Stationary print head / descending build for arbitrarily long rocket bodies
- Porous matrix oxidizer loading without valves or fill lines in propulsion path
- Explosive base ignition of sealed porous reactive metal matrix
- Spin stabilization via helical surface features printed integrally with rocket body
- Mission-specific geometry — each rocket printed with unique AI-optimized structure for target orbit
- Payload module as structural continuation of rocket body, manufactured in same process
| Parameter | Value |
|---|---|
| Length | 12 m |
| Diameter | 0.4–0.5 m |
| Matrix material | Mg alloy AZ31 or similar |
| Oxidizer | LOX |
| Matrix porosity | ~65% |
| Mg matrix mass | ~919 kg |
| LOX mass | ~1,118 kg |
| Total propellant mass | ~2,037 kg |
| Estimated Isp | 280–310 s |
| Target payload | 100–200 kg |
| Target orbit | LEO 400–600 km |
Single-stage orbital capability depends on AI-optimized burn profile and achieved Isp. Two-stage variant — two printed tubes, second stage smaller diameter — is baseline fallback.
| Item | Cost |
|---|---|
| Mg alloy powder LPBF grade ~919 kg × $30/kg | $27,500 |
| LOX ~1,118 kg × $0.12/kg | $135 |
| Machine time ~150 hrs × $10/hr amortized | $1,500 |
| Avionics + fins + initiator | $8,000–12,000 |
| Total per unit (serial production) | $37,000–41,000 |
At scale with in-house powder production: $20,000–25,000 per unit is a realistic target.
| Scenario | Unit Cost | Launch Price | Payload | $/kg LEO |
|---|---|---|---|---|
| Conservative | $40k | $200k | 100 kg | $2,000 |
| Baseline | $30k | $150k | 150 kg | $1,000 |
| Serial production | $20k | $100k | 200 kg | $500 |
| Parameter | Falcon 9 | Rocket Lab Electron | SpinLaunch (Orbital) | CRB |
|---|---|---|---|---|
| Payload to LEO | 22,800 kg | 300 kg | ~200 kg | 100–200 kg |
| Launch Price | ~$70M | ~$8M | < $500k (target) | ~$50–100k |
| $/kg LEO | ~$2,700 | ~$26,000 | ~$2,500 | $500 |
| Parts Count | ~100,000 | ~10,000 | ~1,000s (Excl. Accelerator) | ~1 |
| Assembly Required | Months | Weeks | Days (Vehicle only) | None |
| Moving Parts | Yes (Turbopumps) | Yes (Electric Pumps) | No (Vehicle) / Yes (Accelerator) | Zero |
| Manufacturing | Multi-step manual | Multi-step manual | Composite winding/CNC | Single LPBF print |
| Post-processing | Extensive | Extensive | Moderate | None |
| Human Labor | High | High | Moderate | Near zero |
| Order to Launch | Months | Months | Weeks | Days |
| Reusability | Partial | No | Partial (Accelerator) | Not applicable |
The competitive advantage is not reusability or peak performance. It is radical manufacturing simplicity. The entire production pipeline is:
digital model → LPBF printer → oxidizer loading → launch
No assembly. No post-processing. No moving parts to inspect or certify. Fully automated from file to flight.
Target market: microsatellite operators needing dedicated low-cost launches on short notice — not shared rideshare slots on large vehicles scheduled 18 months out.
- This is not a complete engineering design
- This is not a tested system
- This is a concept disclosure for prior art purposes
- Actual Isp achievable in porous Mg+LOX matrix combustion
- Minimum pore size for stable combustion vs capillary retention of LOX
- MgO slag accumulation effects on thrust profile
- AI optimization convergence for trajectory-encoded geometry
- Numerical simulation of continuously-consuming structural mass (non-standard Tsiolkovsky)
CERN Open Hardware Licence Strongly Reciprocal v2 (CERN OHL-S v2)
Copyright (c) 2026 Aleksei Shtorm / Solenopsys
You may use, copy, modify and distribute this work and derivatives freely, including for commercial purposes, under the following conditions:
- All derivative works must be released under the same CERN OHL-S v2 license
- All modifications must be publicly documented
- You may not use patents or trade secrets to restrict use of this work or its derivatives
Full license text: https://ohwr.org/cern_ohl_s_v2.txt
This publication establishes prior art. Any patent filed after March 10, 2026 covering the above claims is invalid under prior art doctrine.
