1. The short version, what to know first, and where this fits

• Lightness becomes payload. A top-stage structure 100 kg lighter can carry 100 kg more payload (a simple example of ours).Our calculation The Space Shuttle external tank was cut from 76,000 lb to 58,500 lb by changing materials.Sourced[Ref. 1]
• Aluminum alloys, especially lithium-bearing ones, are the mainstay. Lithium is the lightest metal; adding it lowers an alloy's density and raises stiffness. SpaceX says the Falcon 9 first-stage tanks are aluminum-lithium alloy.Sourced[Refs. 7, 13] Titanium alloys serve in high-temperature, high-pressure and cryogenic parts, while magnesium alloys are at the research and validation stage.
• The material and how it is made are one product. Forming, a two-step heat treatment, friction-stir welding, X-ray inspection and cryogenic tests form a chain that protects performance.Sourced[Refs. 5, 4, 9]

2. Why lightness pays

A rocket moves while becoming lighter as it burns fuel. For a stage (the part dropped once its fuel is used), the structure mass (dry mass) counts on the same side as the payload we want to carry. The velocity gain Δv is set by the specific impulse (the "fuel economy" of the propellant) and the ratio of initial to final mass.Sourced[Refs. 24, 23]

In this example, if the structure gets 100 kg heavier with the same payload, Δv falls from about 7,907 m/s to about 7,728 m/s.Our calculation To keep Δv, payload must drop by 100 kg. Conversely, a structure 100 kg lighter allows 10% more payload.Our calculation Structure mass trades almost one for one against cargo, which is why light materials are so valuable.

More detail: the calculation

Multiplying the specific impulse of 450 s by g0 = 9.80665 m/s² gives an effective exhaust velocity of 4,413 m/s. The baseline is ln((1,000 + 10,000 + 1,000) / (1,000 + 1,000)) = ln 6 ≈ 1.7918, so Δv ≈ 7,907 m/s. With 1,100 kg of dry mass, ln(12,100 / 2,100) ≈ 1.7513 and Δv ≈ 7,728 m/s (about 2.3% lower).Our calculation If the sum of dry mass and payload is constant, the content of the logarithm does not change and Δv is kept.

3. A map of light metals by density

In NIST's table, metal densities are 0.534 for lithium, 1.740 for magnesium, 2.699 for aluminum, 4.540 for titanium and 7.874 g/cm³ for iron.Sourced[Ref. 14] Lithium is the lightest metal, lighter than magnesium, at about 0.2 times aluminum.Our calculation Aluminum-lithium alloys add a little of this lithium to make aluminum alloys lighter and stiffer.Sourced[Ref. 7]

Density alone does not decide a material. Specific strength (strength divided by density) and specific stiffness (elastic modulus divided by density) are the gauges of light and strong. For example, the typical yield strength of 2195-T8 (1.85 in plate) is 76 ksi (524 MPa); with a density similar to the 0.098 lb/in³ (2.71 g/cm³) of alloy 2050, the specific strength is about 193 kN·m/kg.Our calculation[Ref. 6]

4. Aluminum alloys: from 2219 to aluminum-lithium

The mainstay of rocket tanks is aluminum alloy, and the Space Shuttle external tank shows its history well. According to a 2011 NASA presentation, the first "standard weight tank" was an aluminum 2219 structure weighing 76,000 lb, the "light weight tank" weighed 66,000 lb and flew 86 missions, and the "super light weight tank" weighed 58,500 lb and was primarily an aluminum-lithium structure.Sourced[Ref. 1] Because the tank flies to orbital velocity with the orbiter, every pound saved can go to payload.Sourced[Ref. 1]

The main material of the super light weight tank is the aluminum-copper-lithium alloy 2195. A 2002 NASA report says that using 2195 on the external tank saved more than 7,000 lb and enabled the launch of International Space Station components.Sourced[Ref. 2] A 1999 NASA presentation says the super light weight tank provided two thirds of the weight savings needed to place the station in a 51.6° orbit.Sourced[Ref. 3]

The lineage continues in later plans. A 2014 NASA paper says 2195 has higher mechanical properties and lower density than 2219, that 2219 is the baseline material for SLS cryogenic tank components, and that replacing 2219 with 2195 would give large weight savings, as on the Shuttle external tank.Sourced[Ref. 5] SpaceX says the Falcon 9 first stage holds liquid oxygen and kerosene (RP-1) in aluminum-lithium alloy tanksSourced[Ref. 13] (company statement; the alloy number could not be confirmed).

The 2050 composition is from the analysis of two plates in NASA TM-2011-217163 (Cu 3.48 to 3.56, Li 0.88 to 0.90 weight percent) and the AMS 4413 range (Li 0.70 to 1.30).Sourced[Ref. 6]

5. Getting stronger in the cold

Tanks for liquid hydrogen and liquid oxygen operate on the cold side. NASA TM-2011-217163 gives values for 2195 and 2050 at room temperature (75 °F) and at −320 °F (about −196 °C). For 2195-T8 (1.85 in plate), typical values in the L direction rise from 76 to 93 ksi in yield strength and from 80 to 101 ksi in tensile strength. In the 2050-T84 (4 in plate) tests, means in the L direction at t/2 rose from 74.6 to 85.9 ksi in yield strength and from 79.2 to 95.7 ksi in tensile strength.Sourced[Ref. 6]

Fracture toughness too is reported to rise for 2050 when temperature falls from room temperature to −320 °F.Sourced[Ref. 6] In contrast to many steels that turn brittle in the cold, aluminum alloys stay strong and tough when cold (general remark; our commentary). This is one reason aluminum alloys are the mainstay of cryogenic tanks.

6. Protecting performance by process: forming, heat treatment, welding, inspection

Tank material is no good as plate. Performance appears only after bending, heat treating, welding and inspecting. Here are four process steps documented by NASA.

View original SVG, values and sources

Original SVG reference: Conceptual diagram (vector drawing). The process chain of an aluminum-lithium tank. The content of each step is in NASA sources.

(1) Forming: stretching and bending gores

The rounded ends (domes) of a tank are welded from curved "gores" (petal-shaped pieces). 2195 was hard to form into thick gores. NASA Marshall developed an annealing treatment that widened the forming range and raised the strain-hardening exponent, and made gores 0.525 in and 0.75 in thick by stretch forming. Tests showed higher than expected strength, modulus and elongation.Sourced[Ref. 5]

(2) Heat treatment: a two-step age

NASA Marshall developed a two-step artificial age for 2195 (20 hours at 132 °C followed by 42 hours at 138 °C) that raises cryogenic fracture toughness and stress-corrosion resistance. It kept the same yield strength, improved short-transverse ductility, and five previously rejected lots passed simulated service testing. The stress-corrosion test required at least 10 days in alternate immersion in 3.5% salt water at 45 ksi, and lives ranged from 57 to 83 days.Sourced[Ref. 4]

(3) Welding: joining without melting

Aluminum-lithium alloys tend to crack and form porosity in conventional fusion welding (general remark). Friction-stir welding softens and mixes the material with a rotating tool and joins it without melting. According to Ding et al., when it was brought to NASA in 1994 it was little more than a laboratory curiosity; the first weld was made in the fall of 1995 and it went into external tank production.Sourced[Ref. 9] For 2195 plate and extrusions, the large forces on the tool were a production challenge that Lockheed Martin Michoud addressed with tooling.Sourced[Ref. 8] Marshall set up a facility able to make longitudinal welds about 5 m long by retrofitting a welding system designed for tank barrels 2 to 10 m in diameter.Sourced[Ref. 10]

(4) Inspection: the "heavy inclusions" that turned up

In friction-stir welds between a 2219 forging and 2195 plate on the external tank, radiography first revealed "heavy inclusions" (HI). They were small particles (about 0.005 to 0.025 in) or clusters, and elemental analysis showed large agglomerations of Al2Cu (theta phase), the strengthening precipitate of 2219. They were not foreign material or tool debris. Areas with HI were tested in room-temperature and cyclic cryogenic (−423 °F) tension and compared with specimens without HI and historical data; no significant adverse effect was found.Sourced[Ref. 11] Non-destructive inspection methods to verify friction-stir welds were also developed for the external tank.Sourced[Ref. 12]

7. Titanium alloys

Titanium is about 1.68 times denser than aluminum (NIST elemental values),Our calculation[Ref. 14] but it is strong and holds up to heat and cryogenic temperatures, so it serves in high-pressure and high-temperature parts. What matters is that compatibility with the partner (propellant, hydrogen) is checked by testing before use.

• Pressure vessels and propellant tanks: forged Ti-6Al-4V was used in Apollo spacecraft propellant tanks (pressure vessels) and its flaw growth was studied (1967).Sourced[Ref. 15] Ti-6Al-4V for satellite propellant tanks was not adversely affected by 2,000 hours in nitrogen tetroxide (MON-1), after 1,000 hours in Freon, at stresses up to 80% of the experimentally found critical value (1983 tests at 49 °C).Sourced[Ref. 16]
• New propellants: for the green propellants AF-M315E and LMP-103S, the threshold stress intensity for environment-assisted cracking (KEAC) of Ti-6Al-4V was measured at 50 °C over 1,000 hours: at least 22.0 ksi√in for base metal and 31.3 ksi√in for welds with AF-M315E,Sourced[Ref. 17] and 22.5 and 36.2 ksi√in with LMP-103S (interim results).Sourced[Ref. 18] These are input values for safe-life analyses of tanks.
• Liquid-hydrogen pumps: Ti-5Al-2.5Sn ELI was used in critical liquid-hydrogen components of the SSME. Titanium's reactivity demanded special compositions and processing.Sourced[Ref. 19] The notch strength of Ti-6Al-4V in low-temperature, high-pressure hydrogen was examined in 1970.Sourced[Ref. 20]
• Spacecraft carrier structure: the Orion heat-shield carrier combines a titanium skeleton with a composite skin.Sourced[Ref. 21]

8. Magnesium alloys

Magnesium is among the lightest structural metals. In NIST elemental values its density is about 0.64 times that of aluminum (1.740 / 2.699).Our calculation[Ref. 14] A 2017 NASA Kennedy Space Center intern report says that although recent magnesium alloy development has widened the potential for spacecraft use, NASA has not been able to consider adoption for lack of test data.Sourced[Ref. 22]

Traditional magnesium alloys tended to be passed over in flight part selection because they are flammable and hard to extinguish and because they are galvanically active and weak in salt water. The report says the recent WE43 and Elektron 21 are described as self-extinguishing and passing an FAA flammability rating, and an anodizing process for magnesium (Tagnite) was formulated to curb galvanic and salt-water corrosion. KSC planned to compare three alloys, AZ31B, WE43 and Elektron 21, against 7075-T6 aluminum and 304 stainless steel for flammability (Test 17 of NASA-STD-6001B), stress corrosion (MSFC-STD-3029), seashore corrosion and mechanical properties.Sourced[Ref. 22]

9. How to read the yardsticks and standards

The performance of lightweight structural materials is measured by yield strength, tensile strength and elastic modulus (at each temperature), fracture toughness (strength with a crack present), resistance to fatigue-crack growth, resistance to stress-corrosion cracking, weld quality, and compatibility with the partner propellant or oxygen. No single number decides; read each with its conditions (temperature, thickness, orientation, environment).

The arrangement of the table is Our calculation (organizing). Numbers, titles and status were checked on ecss.nl and the NASA Technical Standards System (17 October 2026). Standard texts were not read; only titles and scope are given.

Reading notes (for materials engineers)

• Ground tests and flight differ. The values here come from laboratory tests of specific plates by specific methods. Real tanks see welds, thickness variations, repeated filling and pressurization, and environmental effects.
• Typical values and design allowables differ. The 2195 values are "typical" and the 2050 values are "test means". Designs use statistically guaranteed A- or B-basis values, which were not checked here.
• Check whether a standard is a recommendation or a requirement. NASA-STD and ECSS standards become requirements when a project adopts them by contract (general remark). If something claims compliance, check which document and edition.
• Company statements differ from independent verification. That the Falcon 9 tanks are aluminum-lithium is a company statement, and the alloy number could not be confirmed.

10. Glossary

11. References

1. Pilet J. C. et al. (NASA) "External Tank Program - Legacy of Success" (2011) — ntrs.nasa.gov/citations/20120001570
2. Shah S. et al. (NASA Marshall) "Effects of Thermal Exposure on Properties of Al-Li Alloys" (2002) — ntrs.nasa.gov/citations/20030001044
3. Bickley F., Schwinghamer R. J. (NASA) "NASA Experience with the Shuttle External Tank" (1999) (only the abstract of the slides was checked) — ntrs.nasa.gov/citations/19990047392
4. Chen P. S., Stanton W. P. (NASA Marshall) "Cryogenic Fracture Toughness Improvement for the Super Lightweight Tank's Main Structural Alloy" (2002) — ntrs.nasa.gov/citations/20020030745
5. Volz M. P. et al. (NASA Marshall) "Development of Aluminum-Lithium 2195 Gores by the Stretch Forming Process" (2014) — ntrs.nasa.gov/citations/20140011616
6. Hafley R. A. et al. (NASA Langley) "Evaluation of Aluminum Alloy 2050-T84 Microstructure and Mechanical Properties at Ambient and Cryogenic Temperatures", NASA TM-2011-217163 — ntrs.nasa.gov/citations/20160010564
7. Hales S. J. (NASA Langley) "Aerospace Applications of Aluminum-Lithium Alloy Thick Plate" (abstract, 2010) — ntrs.nasa.gov/citations/20200010741
8. Loftus Z. et al. "Friction Stir Weld Tooling Development for Application on the 2195 Al-Cu-Li Space Transportation System External Tank" (1998) — ntrs.nasa.gov/citations/19980201001
9. Ding J. et al. (NASA Marshall) "A Decade of Friction Stir Welding R and D at NASA's Marshall Space Flight Center and a Glance into the Future" (2006) — ntrs.nasa.gov/citations/20080009619
10. Russell C., Ding R. J. (NASA Marshall) "Friction Stir Welding of Large Scale Cryogenic Tanks for Aerospace Applications" (1998) — ntrs.nasa.gov/citations/19990103022
11. Rietz W. W., Jr. "Friction-Stir Welding - Heavy Inclusions in Bi-metallic welds of Al 2219/2195" (2008) — ntrs.nasa.gov/citations/20080036591
12. Suits M. W. et al. (NASA) "Nondestructive Inspection Techniques for Friction Stir Weld Verification on the Space Shuttle External Tank" (2003) — ntrs.nasa.gov/citations/20030112192
13. SpaceX "Falcon 9" (vehicle page) (company statement; checked 17 October 2026) — www.spacex.com/vehicles/falcon-9/
14. NIST "X-Ray Mass Attenuation Coefficients: Table 1, elemental data (density)" (checked 17 October 2026) — physics.nist.gov/PhysRefData/XrayMassCoef/tab1.html
15. Tiffany C. F., Masters J. N. (NASA) "Investigation of the flaw growth characteristics of 6Al-4V titanium used in the Apollo spacecraft pressure vessels" (1967) (only the abstract was checked) — ntrs.nasa.gov/citations/19670015308
16. Bjorklund R. A. "Stress-corrosion crack-growth study of titanium alloy Ti-6Al-4V exposed to freon PCA and nitrogen tetroxide MON-1" (1983) — ntrs.nasa.gov/citations/19830026835
17. Sampson J. W. et al. (NASA) "Fracture Growth Testing of Titanium 6AL-4V in AF-M315E" (2015) — ntrs.nasa.gov/citations/20150022931
18. Mulkey H. W. et al. (NASA) "Fracture Mechanics Testing of Titanium 6AL-4V in LMP-103S Propellant" (2024) (interim results) — ntrs.nasa.gov/citations/20240005074
19. Halchak J. A. et al. "The Development of Titanium Alloys for Application in the Space Shuttle Main Engine" (2010) — ntrs.nasa.gov/citations/20100021913
20. Chandler W. T., Walter R. J. "Low-temperature embrittlement of Ti-6Al-4V and Inconel-718 by high pressure hydrogen" (1970) (only the abstract was checked) — ntrs.nasa.gov/citations/19700000351
21. Koenig W. J. et al. "Orion Heat Shield Manufacturing Producibility Improvements for the EM-1 Flight Test Program" (2019) — ntrs.nasa.gov/citations/20190001310
22. Aroh J. (NASA Kennedy Space Center) "Magnesium Alloys for Space Hardware Design" (intern final report, 2017) (an intern report at the plan and early-data stage) — ntrs.nasa.gov/citations/20170011677
23. Huzel D. K., Huang D. H. "Design of Liquid Propellant Rocket Engines", NASA SP-125 (2nd ed.) (the rocket equation is referenced as a textbook relation) — ntrs.nasa.gov/citations/19710019929
24. NASA Glenn Research Center "Specific Impulse" (Beginner's Guide to Aeronautics) (checked 17 October 2026) — www1.grc.nasa.gov/beginners-guide-to-aeronautics/specific-impulse
25. NASA NASA-STD-6016 "Standard Materials and Processes Requirements for Spacecraft" (Version C with Change 1, change date 15 November 2023) (Active; checked 17 October 2026) — standards.nasa.gov/standard/NASA/NASA-STD-6016
26. NASA NASA-STD-6012 "Corrosion Protection for Space Flight Hardware" (Version A, 23 March 2022) (Active; checked 17 October 2026) — standards.nasa.gov/standard/NASA/NASA-STD-6012
27. NASA Marshall MSFC-STD-3029 "Guidelines for the Selection of Metallic Materials for Stress Corrosion Cracking Resistance in Sodium Chloride Environments" (Version A, 25 February 2005) (Active; checked 17 October 2026) — standards.nasa.gov/standard/MSFC/MSFC-STD-3029
28. NASA NASA-STD-5019 "Fracture Control Requirements for Spaceflight Hardware" (Version A with Change 4, change date 12 July 2024) (Active; checked 17 October 2026) — standards.nasa.gov/standard/NASA/NASA-STD-5019
29. NASA NASA-STD-6001 "Flammability, Offgassing, and Compatibility Requirements and Test Procedures" (Version B with Change 3) (Active; checked 17 October 2026) — standards.nasa.gov/standard/NASA/NASA-STD-6001
30. ECSS ECSS-Q-ST-70-36C "Material selection for controlling stress-corrosion cracking" (6 March 2009) (checked 17 October 2026) — ecss.nl/standard/ecss-q-st-70-36c-material-selection-for-controlling-stress-corrosion-cracking/
31. ECSS ECSS-Q-ST-70-71C Rev.1 "Materials, processes and their data selection" (15 October 2019) (checked 17 October 2026) — ecss.nl/standard/ecss-q-st-70-71c-rev-1-materials-processes-and-their-data-selection-15-october-2019/
32. ECSS ECSS-E-ST-32-01C Rev.2 "Fracture control" (30 July 2021) (checked 17 October 2026) — ecss.nl/standard/ecss-e-st-32-01c-rev-2-fracture-control-30-july-2021/

12. Claim-to-source audit

Last updated: 17 October 2026. Sources are limited to primary materials (NASA reports, presentations and standards, ECSS, NIST, an official SpaceX page). Company announcements are treated as company statements. Numerical examples (Δv, mass differences, density ratios, specific strength) are explanatory calculations made for this article. Design allowables, an international standard for friction-stir welding, flight qualification of magnesium alloys and the alloy number of the Falcon 9 tanks could not be confirmed and are not stated. The magnesium density statement in the NASA KSC intern report disagrees with NIST values, so NIST values were used. Figs. 1 and 8 are AI-generated concept images and do not depict real facilities or products.