---

1. Mission Statement and Driving Requirements

The system shall deliver a self-replicating manufacturing capability to another stellar system, establish an industrial base from local materials, construct and launch daughter probes, and maintain information fidelity across an unbounded number of replication generations.

Five requirements dominate the entire design space:

R1 — Minimum launch mass. The energy cost of interstellar transfer scales linearly with mass while replication capability scales with information, not mass. Everything that can be rebuilt at the destination shall be encoded, not carried. This single requirement collapses the design from a “ship” to a “seed.”

R2 — Deceleration without propellant. The rocket equation is fatal at interstellar velocities (Δv = 0.1c with even ideal fusion exhaust at 0.05c implies mass ratios in the thousands). Both acceleration and deceleration must therefore be propellantless, using energy and momentum sourced from outside the vehicle.

R3 — Survival of a multi-decade to multi-century unpowered cruise through the interstellar medium (ISM), including relativistic dust impacts and an integrated cosmic-ray dose on the order of 10–50 kGy with no possibility of repair resupply.

R4 — Closure of the replication loop. The destination industrial base must be able to manufacture 100% of the probe’s own bill of materials, including its most complex components (computing substrate, sail material, the assemblers themselves), from asteroidal and cometary feedstock. Partial closure is mission failure on generational timescales.

R5 — Generational information integrity. The probability of an uncorrected, propagating error in the constructive genome must be held below ~10⁻¹² per generation. An exponentially replicating system with mutation is an evolutionary system; an evolutionary system sheds its mission constraints. This is treated as a safety-of-mission and safety-of-others requirement (see §9).

2. System Architecture Overview

SRIP-1 is a three-segment architecture. Only the second segment travels.

The Launch Segment is fixed infrastructure in the origin system: a phased-array laser (“beamer”) of order 100 GW optical output, and its power plant. It never leaves and is not part of the replicated payload as hardware — but its blueprint is, because each colonized system must construct its own beamer to launch the next generation (§8).

The Cruise Segment is the seed vehicle itself: a 12 kg spacecraft consisting of a sail, a shield, a frozen manufacturing kernel, and the genome archive. Full mass budget in §3.

The Foundry Segment does not exist at launch. It is what the seed becomes after arrival: a self-expanding industrial ecology on a carbonaceous asteroid, described in §7.

The design philosophy throughout is biological staging: like a plant seed, the vehicle carries a minimal germination kit and a complete genome, and bootstraps everything else from environmental resources. The engineering consequence is that the cruise vehicle contains almost no mechanisms — mechanisms are mass, failure modes, and radiation targets. It is closer to a hardened data archive with a propulsion attachment than to a spacecraft in the traditional sense.

3. Mass Budget (Cruise Segment)

Subsystem	Mass	Notes
Genome archive (×3 redundant)	0.6 kg	§6; includes substrate and shielding overhead
Germination kernel	3.5 kg	Seed assemblers, catalyst inventory, first-light PV film
Cruise computer + clock	0.4 kg	Reversible-logic core, mostly dormant
Erosion/impact shield	2.5 kg	Monolithic graphite, edge-on configuration
Magsail coil (stowed)	3.0 kg	Superconducting loop, doubles as radiator when deployed
Sail (jettisoned after boost)	1.5 kg	Dielectric metasurface, 10⁴ m² class
Structure, harness, margin	0.5 kg
Total	12.0 kg

Twelve kilograms is deliberately conservative relative to gram-scale concepts (Starshot-class). The extra mass buys triple genome redundancy, a real magsail, and a germination kernel large enough to close the replication loop in years rather than centuries. The launch energy penalty is acceptable because the beamer is reusable infrastructure.

4. Propulsion

4.1 Boost: Beamed Lightsail

Acceleration is provided by the origin-system beamer illuminating a dielectric metasurface sail. Kinetic energy at the design cruise velocity of 0.1c is E = (γ−1)mc² ≈ 5.4×10¹⁵ J for 12 kg — about 1.5 TWh, or 100 GW delivered (after all efficiencies) for roughly 40 hours of illumination. This sets the boost distance at several thousand AU, which in turn sets the beamer aperture: diffraction-limited spot maintenance over that range at ~1 µm wavelength requires an effective aperture in the tens-of-kilometres class, realized as a sparse phased array rather than a monolithic optic. Phase coherence across such an array is a control problem, not a physics problem; the required metrology (sub-wavelength distributed phase locking) is an extrapolation of existing techniques.

The sail itself must reflect >99.99% of incident light or it vaporizes: at ~10 kW/m² absorbed, a thin film has no thermal path. Layered dielectric metasurfaces (Si₃N₄/SiO₂ stacks) reach these reflectivities in narrow bands today; the extrapolation is doing it at large area, low areal density (~0.1 g/m²), and with self-stabilizing beam-riding geometry (conical or nested-ring sail shapes that passively re-center on the beam — demonstrated in principle, unsolved in practice).

The sail is jettisoned after boost. Carrying it through the ISM for decades only adds drag cross-section and impact target area.

4.2 Cruise Velocity Selection

0.1c is a compromise, not a limit. Above ~0.15c, ISM dust impact energy (which scales as v²) and the deceleration problem (Δv scales as v, magsail effectiveness improves with v but required braking distance grows) begin to dominate the design. Below ~0.05c, trip times to even nearby targets exceed the demonstrated-confidence window for electronics dormancy. 0.1c gives 44 years to α Centauri and ~120 years to the 12 ly shell, which the architecture treats as the standard hop distance.

4.3 Deceleration: The Actual Hard Problem

Deceleration is where most interstellar concepts quietly die, because there is no beamer at the destination — that asymmetry defines the whole problem for generation zero. SRIP-1 uses a three-stage passive braking chain:

Stage 1 — Magnetic sail (magsail) against the ISM. A deployed superconducting loop (~5 km radius when spun out, current ~10⁴ A) creates an artificial magnetosphere that deflects interstellar protons, transferring momentum to the vehicle. Drag scales favorably at high velocity and weakens as the probe slows — magsails brake efficiently from 0.1c down to ~0.01c over a distance of order 0.5–1 ly and a time of order decades, then become nearly useless. The physics (Zubrin/Andrews) is uncontroversial; the engineering assumption is a high-temperature superconductor with adequate current density that survives 50 years of cosmic-ray displacement damage while operating passively. The braking phase is folded into the flight plan: the final light-year of the trajectory is flown decelerating.

Stage 2 — Electric sail in the stellar wind. Inside the target star’s astrosphere the ISM is replaced by a denser, slower stellar wind. The magsail coil is reconfigured (charged conductor mode, e-sail) to brake against it, scrubbing velocity from ~0.01c down to ~300 km/s during the fall through the outer system.

Stage 3 — Photon and gravity braking at the star. The final capture uses close perihelion passage: radiation pressure on a small re-deployed braking sail plus an Oberth-geometry gravity interaction shapes the trajectory into a bound, highly elliptical orbit with aphelion in the asteroidal zone. For high-luminosity targets (α Cen A/B class) photon braking alone can shed hundreds of km/s (Heller & Hippke showed this closes for milligram craft; at 12 kg it only needs to close for the last few hundred km/s, which it does).

The chain is fully propellantless, satisfying R2. Its cost is time: arrival deceleration adds 30–60 years to every hop. The architecture accepts this without hesitation — a system with an unbounded mission clock trades time for feasibility everywhere it can.

5. Cruise Survival

5.1 Dust

At 0.1c, a 1 µm ISM silicate grain (~10⁻¹⁵ kg) carries ~0.5 J — a rifle-bullet energy density delivered into a micron-scale footprint, converting to a plasma explosion on impact. Statistical expectation for a 12 ly hop is millions of sub-micron impacts and a handful in the 1–10 µm class. Mitigation is geometric and sacrificial: the vehicle flies needle-like, presenting a ~30 cm² frontal cross-section, behind a monolithic pyrolytic graphite spike whose only job is to ablate. 2.5 kg of graphite provides ~2 orders of magnitude more ablation reserve than the expected fluence erodes. Grains above ~50 µm are mission-fatal and unmitigable at reasonable mass; their expected count per hop is <10⁻³, and this residual risk is retired statistically — by launching seeds in small salvos of 3–5 per target (the beamer is reusable; marginal seed cost is trivial for the origin civilization and for every Foundry thereafter).

5.2 Radiation and Dormancy

Galactic cosmic rays deliver dose to everything, shielded or not (shielding 12 kg against GeV protons is not on the table). The design response is threefold. First, minimize what has to work: the cruise computer runs at microwatt duty, waking on a long schedule for housekeeping, trajectory sensing, and archive scrubbing. Second, make the critical asset (the genome) massively error-coded rather than shielded (§6). Third, make the germination kernel chemically robust rather than functionally intact: the seed assemblers are stored as crystallized, catalytically inert solids that tolerate displacement damage, and are refolded/reconstituted at arrival using the archive as reference — the biological analogy is spore formation, and it converts a radiation-hardness problem into an information problem, which is the kind SRIP-1 knows how to win.

6. The Genome: Generational Information Integrity

This subsystem, not the nanotechnology, is the heart of the design. R5 demands error rates that make ordinary digital storage look like wet clay.

Substrate. The archive is written in a solid-state medium with atomic-scale bits — the reference implementation is dopant-atom placement in diamond lattice (areal density ~10²⁵ bits/kg theoretical; SRIP-1 assumes a derated 10²⁰ bits/kg engineering density). The full constructive genome — every blueprint, every process recipe, the mission constitution (§9), and the beamer design — is estimated at 10¹⁵–10¹⁶ bits, occupying grams. The 0.6 kg allocation buys three physically separated full copies, each internally redundant.

Coding. Data is protected by nested error-correcting codes (inner LDPC, outer Reed–Solomon–class block codes) with combined overhead ~40%, sized so that the uncorrectable word probability over a 150-year hop at the expected upset rate is below 10⁻¹⁸ per copy. Cross-copy majority voting and periodic scrubbing during cruise wake-ups push the end-to-end genome corruption probability per hop below 10⁻¹⁵ — three orders of margin against R5.

Copy discipline. At replication time, the daughter genome is not copied from the working (possibly patched, possibly damaged) Foundry data — it is copied from the master archive, verified against cryptographic hashes that are themselves part of the archive, in a copy-verify-commit protocol analogous to DNA polymerase proofreading but with digital exactness. Any Foundry-local engineering adaptations (and there will be many — every asteroid is different) live in a separate, clearly bounded epigenome that daughters may consult but that carries no authority over the constitution. This separation is what keeps three billion years of necessary local adaptation from becoming three billion years of drift.

7. Arrival and the Foundry: Closing the Replication Loop

7.1 Site Selection and Germination

The bound arrival orbit is trimmed (photon-pressure steering on the residual braking sail) toward a carbonaceous near-star asteroid in the 1–10 km class: C-type bodies offer, in one package, metals (Fe/Ni, often as free metal grains), volatiles (water, CO₂, N, organics), silicates, and milligravity that makes material handling nearly free. The seed anchors, unrolls its first-light photovoltaic film (~10 m², carried), and begins the germination sequence: reconstitute the assembler kernel, verify it against the genome, and set it to work on the one task that matters — building more manufacturing capacity.

7.2 The Industrial Chain

The Foundry’s process chain is deliberately unexotic at the front end and exotic only at the back end. Bulk operations use extrapolated conventional chemistry: solar-thermal volatile extraction, molten regolith electrolysis for oxygen and raw metal (energy cost of order 10 kWh/kg refined metal), carbonyl (Mond) process for high-purity Fe/Ni, carbothermal reduction for silicon. None of this requires molecular manufacturing; it requires reliability in vacuum and milligravity, which is an engineering-maturity assumption, not a physics one.

Molecular manufacturing — machine-phase, positionally controlled mechanosynthesis — is reserved for the components that genuinely need atomic precision: the computing substrate, the genome media, sail metasurfaces, superconductors, and the assemblers themselves. The theoretical basis (Drexler-class analysis, since refined by DFT studies of specific tool-tip reactions on diamond and silicon surfaces) shows no thermodynamic prohibition: reaction site error rates below 10⁻¹⁵ are achievable in principle because each placement can be an individually reversible, checkable event, paying the Landauer cost (kT ln 2 ≈ 3×10⁻²¹ J at 300 K) per verified bit of placement information — an energy overhead that is negligible against the chemical bond energies involved. Throughput comes from convergent assembly: molecular mills build micro-blocks, micro-assemblers build milli-blocks, and so upward, so that macroscale production rate is set by the top of the hierarchy, not the atom-by-atom bottom.

Assumed system-level replication performance: the Foundry doubles its own manufacturing mass every ~6 months in the early phase (consistent with published theoretical self-replication studies that put minimal-system doubling times in the weeks-to-months range; SRIP-1 derates heavily). From a 3.5 kg kernel, ten years of doublings yields a kiloton-class industrial base — more than sufficient for phase three.

7.3 What the Foundry Builds

In priority order: (1) itself, to the kiloton scale; (2) the observatory — because the next generation’s target selection deserves better data than the parent had; (3) the beamer, a replica of the origin launch infrastructure, powered by a growing field of manufactured thin-film PV (a 100 GW beamer at 20% end-to-end efficiency needs ~500 GW of collection, roughly 10³ km² of 0.1 g/m² film — about 100 tonnes of sail-grade material, i.e., weeks of mature-Foundry output); (4) a salvo of daughter seeds; and (5) the long-term local program, whatever the constitution directs (§9). Steps 3–4 satisfy R4’s most subtle clause: the replicated unit is not the probe, it is the probe-plus-launcher system. A von Neumann probe that cannot rebuild its own launch infrastructure is a one-generation firework.

8. Communications

Each Foundry constructs an optical/mm-wave phased array for inter-node links. Direct links across 10 ly are unremarkable at Foundry power levels (a 10 kW optical transmitter through a 100 m synthetic aperture closes a Mbit/s link over 10 ly with modest coding margin). The architecturally interesting option is the solar gravitational lens relay: each system’s star focuses light from ~550 AU outward along the anti-target axis with enormous effective gain; a Foundry-manufactured relay terminal parked on the focal line of each neighbor direction turns the colonization network into a chain of star-lensed trunk links with orders-of-magnitude link-budget margin. The network’s purpose is not command and control — light-lag forbids it, and the architecture is deliberately autonomous — but genome checksum comparison across lineages (a distributed immune system against drift, directly supporting R5) and science return toward the origin.

9. Constitutional Constraints (Safety and Population Control)

An unbounded exponential replicator is the single most dangerous artifact this architecture could produce, so the constraint system is designed with the same rigor as the propulsion. Three mechanisms operate in depth. First, hard replication gating: daughter-genome copying is cryptographically enabled only by completion of a constitution-mandated checklist (target-system survey, biosignature scan, quota verification), with the enabling keys stored in the same triple-redundant archive as everything else — the machine cannot want around it any more than it can want around the ECC. Second, quotas: a per-system cap on converted mass (parts-per-billion of small-body inventory) and a per-generation cap on daughters (3–5), keeping the expansion wavefront information-limited rather than resource-limited. Third, biosphere interdiction: any system flagged with biosignatures or technosignatures shifts the local program from industry to observation under a strict non-interaction envelope — the Foundry parks, watches, and reports.

10. Performance Envelope

Per-hop timeline: boost (days) + cruise at 0.1c (~120 y for 12 ly) + braking (~40 y) + Foundry buildup to launch capability (~15 y) ≈ 175 years per 12 ly generation, an effective expansion speed of ~0.07c including all stops. Sweeping the galactic disk (~10⁵ ly) from one origin therefore takes of order 1.5–2 million years — a number worth staring at, because it is 0.01% of the age of the galaxy. The architecture’s most important output is not any single subsystem but this ratio: within known physics, at technology levels that violate no theoretical bound, a single launch event saturates a galaxy in a geological eyeblink. Total origin-civilization investment: one beamer, one seed salvo, and a constitution they are willing to see enforced, unmodified, ten million times.

11. Technology Gap Register

For honesty, the load-bearing assumptions ranked by distance from current capability: (1) machine-phase mechanosynthesis at system scale — theoretically grounded, experimentally embryonic (single-atom placement demonstrated; mills and convergent assembly are not); (2) 150-year autonomous dormancy with self-repair — no empirical base beyond ~50-year spacecraft; (3) multi-decade passive superconducting magsail operation; (4) 10-km-class phase-coherent laser arrays; (5) sail metasurfaces at 0.1 g/m² with 99.99% reflectivity at large area. Nothing on this list touches a physical law. Everything on it touches manufacturing precision, materials maturity, and verification of extreme reliability — which is to say, the gaps are all in the discipline you and I already practice, just several turns of the crank further on.

---

On method: the underlying ideas and concepts are genuine Nolle Engineering; the detailing and write-up were carried out with AI. Part of this experiment is probing the quality and performance of these tools against real engineering thinking. See the series introduction.