Reusable upper stages promise lower launch costs and faster cadence, but multiple interdependent engineering challenges have so far limited their adoption. Thermal protection during reentry, propellant management in orbit, and the mass penalties of recovery hardware reduce the performance and economic case for a reusable second stage. Elon Musk SpaceX has framed these as central hurdles for making fully reusable architectures routine, and William H. Gerstenmaier NASA has emphasized the tradeoffs between reusability and payload fraction in public commentary.
Thermal and propellant challenges
An upper stage returning from orbital velocity faces much greater heating than a first stage that reenters at suborbital speeds. Designing robust heat shielding that is light enough to preserve payload capability is difficult because added shielding increases structural mass and lowers payload mass fraction. Storing cryogenic propellants for the duration of an orbital mission introduces further engineering burdens. Cryogenic boil-off and the need for active or passive thermal control systems complicate long coast phases and can require additional propellant or insulation that further reduces useful payload unless zero-boil technologies are used, an area of ongoing research at NASA.
Structural, operational, and economic barriers
Reusable upper stages must survive high dynamic loads, thermal cycles, and multiple engine restarts while remaining inspectable and refurbishable. Integrating landing or capture systems adds structures and actuators that become recurring maintenance points. Engines designed for restart and throttling in vacuum add complexity relative to expendable designs. Even if technically feasible, the operational tempo, inspection protocols, and workforce needed for rapid turnarounds create logistic overheads that influence whether reuse reduces life-cycle cost. Cultural and regulatory factors matter as well: countries and companies with different industrial bases and safety cultures will assess acceptable refurbishment risk differently, affecting adoption across territories.
The consequence is that many providers favor reusing the first stage, where return demands are less severe, while treating upper stages as expendable or focusing on in-orbit refueling to decouple reuse from atmospheric recovery. Unless breakthroughs reduce the mass and complexity penalties of thermal protection and cryogenic storage, or new business models that value in-space refurbishment emerge, widespread reusable upper stages will remain an aspirational but costly engineering choice.