The real bottleneck
At Cadence, Propellant Logistics Beats Vehicle Count
Each launch is thousands of tonnes and hundreds of truckloads. Scale that to daily flights and the trucks run out before the pads do.
A Starship launch loads roughly 5,000 t of propellant, about 78% of it liquid oxygen by mass, the rest liquid methane. Each launch currently needs more than 200 tanker trucks of liquid oxygen and nitrogen delivered to the pad. That is manageable at a few flights a year and impossible at the cadence SpaceX is targeting, so the propellant chain is being rebuilt: make the oxygen on site, source the methane next to the gas, and hold a strategic buffer in the tank farm. Beyond Earth orbit the same logic returns as the depot — a place to store propellant that tankers deliver.
This page collects the whole chain; the site pages link back to it. Building
The methalox split
Oxygen Is the Heavy Half; Methane Is the Bought-In Half
The two propellants have completely different supply stories.
Liquid oxygen is about 78% of the load and can be made from air anywhere with enough power, which is why SpaceX is bringing oxygen and nitrogen production on site. Liquid methane is roughly 22% and is bought as commercial liquefied natural gas: Raptor is designed to burn it without an on-site pretreatment plant, so SpaceX dropped its planned gas pretreatment and liquefier in 2022 once Raptor 2 proved commercial LNG worked. That single decision is why methane still arrives by truck while oxygen is going local — and why Louisiana, sited next to Gulf Coast natural gas, changes the methane story too.
Ground supply
From Trucked-In to Made-On-Site
The tank farm, the air separation unit, the Linde plant and the LNG chain.
| Element | What it does | Status | Rated |
|---|---|---|---|
| Starbase orbital tank farm | Stores LCH₄, LOX, LN₂ for subcooling and deluge water; subcoolers chill propellant below boiling for density | Operating; two new methane tanks added in 2026 for dual-pad work | Observed |
| Starbase air separation unit | Separates oxygen, nitrogen and argon from air beside Highway 4 | Under construction since 23 July 2025; ~159 ft tall. Cuts the 200-plus trucks per launch by roughly 75% | Building |
| Linde plant, Brownsville | Industrial-gas supplier producing oxygen, nitrogen and argon on a 72-acre site | Expected online in 2026; cuts delivery from about 10 hours to under 2 | Building |
| Methane (LNG) supply | Commercial liquefied natural gas, trucked in | On-site pretreatment plant and liquefier dropped in 2022 after Raptor 2 proved it burns commercial LNG | Observed |
| Florida tank farms | LC-39A tank farm and flame trench built alongside the pad | Nearing completion with the pad | Building |
| Louisiana propellant production | On-site plant and power generation next to Gulf Coast natural gas | Planned; local gas was a stated reason for the site | Projected |
SpaceX first planned its own air separation plant at the Sanchez site and then at the launch site, abandoned both over power supply and pad proximity, which is why Linde stepped in at Brownsville — and then began the on-pad ASU anyway. The two now overlap, and the ASU’s exact role relative to Linde is not fully settled.
The orbital depot
The Supply Chain Extends Into Orbit
Beyond LEO, propellant supply is the depot, and it has not flown.
Everything above is Earth-side. The moment a mission goes beyond low Earth orbit, propellant supply becomes an orbital problem: a depot filled by tanker flights, holding cryogenic methane and oxygen against boil-off. Ship-to-ship cryogenic transfer — the maneuver that fills the depot — is the single gate on every Moon and Mars variant and has not been demonstrated. It has its own page, Tanker, Depot & Propellant Transfer.
Contested
Deliberately Not Claimed
The load-bearing uncertainties, tagged with the confidence they actually earn.
The ASU-versus-Linde split is unsettled. SpaceX building an on-pad air separation unit while Linde builds a plant to supply the same commodities leaves the division of supply unclear from outside. Both are under construction; which carries the base load is not stated.
The 5,000 t figure is the full stack load. It combines booster and ship and varies by block — Block 3 loads more than Block 2, Block 4 more again. Treat it as an order-of-magnitude anchor, not an exact per-flight number.
Louisiana’s local-methane advantage is a plan. Gulf Coast gas is why the site was chosen, but the on-site production facility is announced, not built.
Record
Provenance and Data Sources
- Confidence
- Observed for the operating tank farm and LNG chain — Building for the ASU and Linde — Projected for Louisiana
- Status date
- Q3 2026
- Ground supply
- Cameron County air separation unit approval (July 2025); FAA 2022 Starbase programmatic assessment; Linde Brownsville plant reporting
- Methane
- 2022 FONSI removing the gas pretreatment plant and liquefier
- Weakest link
- The ASU/Linde division of supply, and the unflown orbital depot