Source Review: New Space Economy "Orbital Refueling of the Starship Architecture" (Dec 2025)

Summary

Claim 1: 10-20 tanker launches per HLS mission, mid 12-14

Quote: "Current estimates range from '10 to nearly 20 launches'... 'High single digits to high teens (e.g., 15-18 launches)' typical" Verdict: Consistent Why: My calc's gear-ratio amplification factor (~12-14×) tracks this. Captured in q2.c12.

Claim 2: ~1,200 tons propellant per Starship HLS mission

Quote: "a fully fueled Starship HLS requires approximately 1,200 tons of propellant" Verdict: Consistent Why: With 100 t lunar payload, gives Γ ≈ 12 — matches Metzger's Γ_LEO ≈ 14 within tolerance. The 14 vs 12 difference is roughly the return-leg propellant overhead.

Claim 3: ~$400M per HLS mission

Quote: "the price is likely to be $400M per run (including HLS investment recapture and 50% profit over 10 uses)" Verdict: Consistent Why: Direct anchor for q2.c12. $400M / 100 t = $4,000/kg of lunar-surface payload — matches my Earth-imports-only mid-era $4,162/kg within 5%. Strong cross-check.

Claim 4: HLS depreciated over 10 uses, 50% profit margin

Quote: "including HLS investment recapture and 50% profit over 10 uses" Verdict: Novel supporting Why: Provides the reuse-count + margin assumption that q2.c9 captures. The 50% profit margin is a market-pricing assumption that adds about $1,500/kg on top of the internal cost. My calc's hardware $/kg/reuse figures aim at internal cost, not list price; this is consistent.

Anti-hallucination check

All quotes appear in extract.md. This is trade press, not peer-reviewed, but the specific tanker-count and propellant-mass figures are consistent with the SpaceX HLS architecture documentation across multiple sources.

New Space Economy — orbital refueling architecture details

Trade-press analysis of the Starship orbital refueling architecture as of December 2025. Used for tanker-count and propellant-mass figures.

tanker-flight-counts

• "Current estimates range from '10 to nearly 20 launches'" per single Moon mission
• "High single digits to high teens (e.g., 15–18 launches)" typical
• Variation depends on transfer efficiency and boil-off rates

propellant-mass

• "a fully fueled Starship HLS requires approximately 1,200 tons of propellant"
• "each Tanker delivers 100 tons" of propellant to the depot per flight
• Minimum 12 launches before accounting for losses

per-mission-cost-claim

From secondary summary: "the price is likely to be $400M per run (including HLS investment recapture and 50% profit over 10 uses)" with "8–10 tanker runs to the LEO fuel depot to refuel a Cargo Starship headed to Lunar Orbit."

relevance-to-q2

Two anchors:

1. Propellant-mass-to-payload ratio: 1,200 t propellant in LEO supports a 100 t payload delivery to lunar surface, plus return. Approximate gear ratio = 12:1 for one-way payload delivery (matches Metzger's Γ_chemical framework).
2. Per-mission cost: $400M for ~100 t lunar payload delivery → ~$4,000/kg of payload delivered to lunar surface or returned from lunar surface, all-in. This is the Earth-imports-only chemical scenario for q2.

reconciling-with-q1-and-coutts-sowers

If q1's partial-mid Starship internal cost is $107/kg to LEO, then 12 tanker flights × 100 t × $107/kg = $128M propellant cost in LEO per HLS mission. Add HLS hardware amortization (depreciate $2.89B contract over 10 uses → $289M per mission) and the per-mission cost is closer to $400–600M, matching the trade-press anchor. The cost-per-kg of lunar-delivered payload is then $4,000–6,000/kg under chemical-only, no-ISRU architecture in mid-era Starship economics.

limitations

• Trade-press, not peer-reviewed
• Tanker-count range is wide (10–20)
• The $400M/mission figure is a single industry estimate; vintage uncertain
• Doesn't separately model the return-to-LEO leg, which is where lunar ISRU propellant changes the calculus most dramatically