How do we know what the Earth's core is made of?

The core is physically out of reach, so its composition is worked out indirectly. Earthquake waves reveal its structure and density, meteorites and the Sun suggest what the planet was built from, and laboratory experiments test which materials behave the right way under extreme pressure. Together these point to an iron-nickel alloy with some lighter elements mixed in. Exactly which lighter elements, and how much, is still debated. [1] [2]

Isaac Newton calculated roughly three centuries ago that the Earth's average density is about twice that of surface rocks, so the interior must be made of much denser material. His estimate has barely changed. The core itself was discovered in 1906 by R.D. Oldham from earthquake records. The outer core is presumed liquid because it does not transmit shear waves and because compressional waves slow sharply as they pass through it. The inner core is considered solid because of how both wave types behave in it. Seismic observations confirmed this in the 1970s, about 30 years after mineral physics arguments had proposed it. [1] [2]

The wave speeds also constrain composition. A pure iron-nickel mixture would be too dense to match the measured seismic velocities, so a smaller fraction of a lighter element is needed. [3]

Elemental abundances in the Sun and in chondritic meteorites, which are considered remnants of the material that built the planets, are used to estimate the bulk composition of the Earth. Metallic meteorites are also thought to be representative of the core, and they show a ratio of about 85 percent iron to 15 percent nickel, matching calculations for the core. The existence of Earth's magnetic field supports a metallic core as well, since rock is not magnetic. Dynamo theory for the magnetic field is one of the data sources used to identify the core's makeup. [3] [4] [1]

Researchers agree the core is mainly iron and nickel, but the identity and amount of the lighter ingredients are unsettled. A USGS summary describes the core as principally iron with about 10 percent of oxygen, sulfur or nickel, or some combination. A 1997 PNAS paper puts the outer core at about 80 percent iron and 5 percent nickel, with up to 15 percent of a lighter element, and lists oxygen, silicon and carbon as likely candidates, with hydrogen or nitrogen less so. A 2023 review in Nature Communications calls the quantities of lighter elements still debated. The figures differ between sources, which reflects this uncertainty. [1] [3] [2]

Scientists test candidate mixtures in the laboratory at pressures around 2 million times atmospheric pressure and temperatures as high as 2000 degrees Celsius, according to the USGS. Seismology is also improving. A newer technique cross-correlates the late seismic records that follow large earthquakes, which has produced better estimates of inner core shear properties. Those properties help narrow down which forms of iron could make up the inner core. Current inner core models still carry large uncertainties. [1] [2]