1.4 Bulk Earth composition

Composition of Earth's layers

Earth's layered structure is the product of differentiation: the separation of an initially more homogeneous body into chemically distinct shells. Each layer's composition reflects a combination of the raw materials Earth accreted from and the physical-chemical processes that sorted those materials by density and bonding affinity.

Geochemical analysis of each layer feeds directly into models of bulk Earth composition, since no single layer can be sampled in isolation to represent the whole planet.

Crust composition

The crust is Earth's thinnest and most accessible layer, dominated by silicate minerals built from lighter elements.

• Oceanic crust: basaltic composition, roughly 7 km thick, formed at mid-ocean ridges
• Continental crust: more felsic (granitic) composition, averaging ~35 km thick, with a much longer and more complex formation history
• Major element abundances (by mass): O (46.6%), Si (27.7%), Al (8.1%), Fe (5.0%)
• Trace elements tend to concentrate in the crust because they are incompatible in common mantle minerals, meaning they preferentially enter the melt phase during partial melting and get carried upward

Mantle composition

The mantle makes up ~67% of Earth's mass and is composed of denser, Mg-Fe-rich silicate minerals.

• Upper mantle mineralogy: olivine, pyroxenes, and garnet (collectively forming a rock type called peridotite)
• Lower mantle mineralogy: high-pressure phases, primarily bridgmanite (Mg-silicate perovskite) and ferropericlase (magnesiowüstite)
• Major element abundances (by mass): O (44.8%), Mg (22.8%), Si (21.5%), Fe (5.8%)
• These compositions are inferred indirectly from mantle xenoliths (fragments carried up by volcanic eruptions), seismic wave velocities, and high-pressure laboratory experiments

Core composition

The core accounts for ~33% of Earth's mass and is overwhelmingly metallic.

• Divided into a liquid outer core and a solid inner core
• Primarily iron (~85%) and nickel (~5%)
• The core's measured density is lower than a pure Fe-Ni alloy would be, so light elements (O, S, Si, C, H) must be present to account for this density deficit
• Composition is constrained by seismic data, comparison with iron meteorites, and high-pressure experiments
• Temperature at the core-mantle boundary is estimated at ~3800 K

Major element abundances

Six elements make up over 90% of Earth's mass. Knowing how they distribute across layers is central to any bulk composition model.

Oxygen and silicon dominance

• Oxygen is the most abundant element in the bulk Earth by mass (~30%), and even more dominant in the silicate portion
• Silicon is second (~15% of bulk Earth)
• Together, O and Si form the tetrahedral $SiO_4$ units that are the structural backbone of all silicate minerals in the crust and mantle
• The O/Si ratio varies between layers because differentiation concentrates different mineral assemblages at different depths
• These abundances broadly reflect the composition of the solar nebula from which Earth formed

Iron and magnesium content

• Iron is the most abundant element in the core (~85%) but only ~5-6% of the mantle by mass. Its concentration in the core reflects iron's siderophile (metal-loving) character: during differentiation, iron preferentially partitioned into the metallic liquid that sank to form the core.
• Magnesium is the dominant cation in mantle minerals (~22.8% of the mantle), residing mainly in olivine $(Mg,Fe)_2SiO_4$ and pyroxenes
• The Fe/Mg ratio in mantle rocks controls their density, solidus temperature, and seismic velocity, making it a key parameter in geophysical models

Minor elements distribution

• Aluminum: concentrated in the crust (lithophile behavior), a major component of feldspars and clay minerals
• Calcium: important in both crustal minerals (plagioclase, calcite) and mantle minerals (clinopyroxene, garnet)
• Sodium and potassium: strongly enriched in the continental crust due to their large ionic radii, which make them incompatible in dense mantle phases
• Titanium: present in accessory minerals (ilmenite, rutile) throughout the crust and mantle
• Phosphorus: distributed between mantle, crust, and potentially the core (it has some siderophile tendency at high pressures)

Trace element distribution

Trace elements occur at concentrations below 0.1 wt%, but their geochemical behavior makes them powerful diagnostic tools. Because they partition unevenly during melting and crystallization, their distribution patterns record the history of differentiation, melting, and recycling within the Earth.

Lithophile elements

Lithophile ("rock-loving") elements preferentially bond with oxygen and concentrate in silicate minerals.

• Include the rare earth elements (REEs), U, Th, and K
• Strongly enriched in the crust relative to the mantle because they are incompatible during mantle melting
• Used as tracers for partial melting, fractional crystallization, and crustal growth
• U, Th, and K are the main sources of radiogenic heat inside the Earth, so their distribution directly controls the planet's thermal budget and drives mantle convection

Siderophile elements

Siderophile ("iron-loving") elements have a strong affinity for metallic iron.

• Include the platinum group elements (PGEs: Os, Ir, Ru, Rh, Pt, Pd), gold, and nickel
• Concentrated in the core during planetary differentiation
• Mantle abundances of PGEs are far lower than chondritic values, but still higher than metal-silicate partitioning experiments predict. This excess is explained by the late veneer hypothesis: a small amount of chondritic material was added to the mantle after core formation was essentially complete.

Chalcophile elements

Chalcophile ("sulfur-loving") elements tend to combine with sulfur.

• Include Cu, Zn, Pb, and Hg
• Concentrated in sulfide minerals, which is why many economically important ore deposits are sulfide-hosted
• Their distribution was affected by both core formation (sulfur likely partitioned partly into the core) and later mantle melting
• Some chalcophile elements (Cu, Zn) also exhibit lithophile behavior in the silicate mantle, so their classification is context-dependent

Isotopic composition

Isotopic ratios are among the most precise tools in geochemistry. They record information about Earth's age, the timing of differentiation events, and the cycling of material between reservoirs.

Stable isotopes in Earth

Stable isotopes do not undergo radioactive decay, but their ratios shift through mass-dependent fractionation during physical and chemical processes.

• $\delta^{18}O$: fractionated by temperature-dependent processes; used to study paleoclimate, water-rock interaction, and magma sources
• $\delta^{13}C$: traces carbon cycling between the atmosphere, biosphere, ocean, and mantle
• $\delta^{34}S$: sensitive to redox conditions and biological sulfur metabolism

Radiogenic isotopes

Radiogenic isotopes are produced by the decay of radioactive parent nuclides over time.

• Key decay systems: $^{87}Rb \rightarrow ^{87}Sr$, $^{147}Sm \rightarrow ^{143}Nd$, $^{238}U \rightarrow ^{206}Pb$, $^{176}Lu \rightarrow ^{176}Hf$
• Used for geochronology (dating rocks and minerals) and as tracers for mantle and crustal processes
• Ratios like $^{87}Sr/^{86}Sr$ and $^{143}Nd/^{144}Nd$ differ between mantle reservoirs because parent-daughter fractionation occurred at different times and to different degrees, creating isotopic "fingerprints" for each reservoir

Bulk silicate Earth model

The Bulk Silicate Earth (BSE) represents the combined composition of the mantle plus crust, essentially everything except the core. It's the most widely used reference frame for discussing Earth's silicate composition.

Definition and assumptions

• Assumes the mantle and crust were homogeneously mixed immediately after core separation, before any crust extraction occurred
• The core is excluded because it cannot be directly sampled
• BSE composition is estimated from analyses of mantle-derived rocks (peridotites, basalts) and comparisons with meteorites
• Refractory lithophile elements (those that condense at high temperatures and prefer silicates) are assumed to occur in chondritic ratios, since no known process fractionates them from each other during planetary formation
• Volatile elements are recognized as depleted relative to chondrites

Compositional estimates

• Major elements (wt%): Si ~21%, Mg ~22%, Fe ~6%, Al ~2.3%, Ca ~2.5%, Na ~0.3%
• Trace element abundances are typically normalized to CI chondrite values to reveal enrichment/depletion patterns
• The BSE REE pattern is relatively flat when normalized to chondrites, consistent with the chondritic assumption for refractory lithophile elements
• Highly siderophile elements are strongly depleted (extracted into the core)
• Moderately volatile elements (K, Rb, Cs) are depleted relative to CI chondrites, reflecting incomplete accretion of volatile-bearing material

Primitive mantle composition

The primitive mantle is conceptually identical to the BSE: it represents the mantle's composition right after core formation but before any crust had been extracted. It serves as the baseline against which all subsequent differentiation is measured.

Chondritic Earth hypothesis

The standard model assumes Earth's building blocks had bulk compositions similar to CI carbonaceous chondrites, the most chemically primitive meteorites.

• Predicts that refractory lithophile element ratios (e.g., Ca/Al, Sm/Nd) in the primitive mantle should match CI chondrites
• Supported by near-identical isotopic compositions of O, Cr, and Ti between Earth and certain chondrite groups
• Successfully explains siderophile element depletions as a consequence of core formation
• Challenge: Earth's Mg/Si and Al/Si ratios don't perfectly match any single chondrite group, suggesting either the building blocks were a mix of materials or that some process (like collisional erosion) modified these ratios

Non-chondritic models

Several alternative models have been proposed to address the mismatches:

• Enstatite chondrite model: proposes Earth formed from reduced, volatile-poor material with isotopic compositions closer to Earth's for O, Cr, and Ti
• Collisional erosion model: suggests early-formed crust (enriched in Al, Si) was stripped away by giant impacts, shifting bulk ratios
• Mixed-source models: invoke accretion from a blend of different chondritic and non-chondritic materials
• These models remain actively debated, and no single model satisfies all geochemical and isotopic constraints simultaneously

Core formation effects

Core formation was the most significant chemical differentiation event in Earth's history, redistributing elements between the metallic core and the silicate mantle based on their metal-silicate partition coefficients.

Siderophile element depletion

• During core formation, siderophile elements partitioned strongly into the segregating metal phase
• The degree of depletion in the mantle scales with each element's metal-silicate partition coefficient ($D_{met/sil}$): higher $D$ means stronger depletion
• Moderately siderophile elements (Ni, Co) are depleted by factors of ~10-20 relative to chondrites
• Highly siderophile elements (PGEs, Au) are depleted by factors of ~100-1000, yet their mantle abundances are still higher than equilibrium partitioning predicts
• This excess is best explained by a late veneer: ~0.5% Earth mass of chondritic material added after core formation ceased

Light element enrichment

The core's density is 5-10% lower than a pure Fe-Ni alloy at core pressures, requiring the presence of light elements.

• Candidates: S, O, Si, C, and H
• Estimated total light element content: 5-10% by mass
• Which light elements dominate remains debated and depends on the pressure, temperature, and oxygen fugacity during core formation
• The partitioning of these light elements between core and mantle affects mantle abundances of Si, O, and S, creating a feedback between core composition models and BSE estimates

Volatile element depletion

Earth is significantly depleted in volatile elements (those with low condensation temperatures) compared to CI chondrites. This depletion is one of the most important constraints on how and where Earth formed.

Causes of depletion

Multiple processes likely contributed:

1. Incomplete condensation: volatile elements may not have fully condensed from the hot solar nebula at Earth's orbital distance
2. Impact-driven loss: high-energy collisions during accretion could have vaporized and ejected volatile-rich material
3. Hydrodynamic escape: early atmospheric loss driven by intense solar radiation
4. Core sequestration: some nominally volatile elements (S, C) may have partitioned into the core rather than being lost entirely

The observed depletion pattern, where elements are progressively more depleted with decreasing condensation temperature, suggests incomplete condensation was a major factor.

Implications for Earth formation

• Earth likely formed primarily from volatile-poor materials in the inner solar system
• The origin of Earth's water and other volatiles remains debated: options include indigenous accretion from hydrated silicates, delivery by carbonaceous chondrite-like bodies, or a combination
• Volatile depletion patterns constrain the temperature and redox conditions of the disk material from which Earth accreted
• These patterns also set boundary conditions for models of early atmospheric composition and the timing of ocean formation

Compositional evolution

Earth's composition has not remained static. Ongoing differentiation, recycling, and exchange between reservoirs have continuously modified the chemistry of each layer over 4.5 billion years.

Differentiation processes

• Core formation (largely complete by ~30-60 Myr after solar system formation) concentrated siderophile elements in the core
• Mantle melting and crust extraction progressively depleted the upper mantle in incompatible elements while building the continental crust
• Fractional crystallization within magma chambers produced the evolved (silica-rich) compositions characteristic of continental crust
• Metamorphism altered rock compositions through mineral reactions and fluid-mediated element transport
• Subduction recycles crustal material back into the mantle, partially reversing the depletion caused by melt extraction

Crust-mantle exchange

• Partial melting of the mantle at ridges and hotspots extracts incompatible elements into basaltic magmas, depleting the residual mantle
• Delamination of dense lower crust returns mafic material to the mantle
• Subduction of oceanic crust and sediments introduces chemical heterogeneities into the mantle, creating isotopically distinct domains
• Fluid and melt migration through the mantle wedge above subduction zones redistributes mobile elements (e.g., Rb, Ba, Pb)
• Mantle plumes may sample deep, relatively primitive mantle, bringing material to the surface that has been isolated from recycling for billions of years

Geochemical reservoirs

Earth's interior is not chemically uniform. Distinct geochemical reservoirs have developed through billions of years of differentiation and recycling, and they can be identified by their characteristic trace element and isotopic signatures.

Depleted mantle

• Mantle that has lost incompatible elements through repeated episodes of melt extraction
• Characterized by low concentrations of incompatible elements (Rb, U, Th, light REEs)
• Isotopic signatures: low $^{87}Sr/^{86}Sr$ (because Rb, the parent, was extracted) and high $^{143}Nd/^{144}Nd$ (because Nd, the daughter element's parent Sm, is less incompatible than Nd itself, so the Sm/Nd ratio increased in the residue)
• This is the source of mid-ocean ridge basalts (MORBs), the most voluminous volcanic product on Earth
• Occupies much of the upper mantle, with portions possibly extending into the lower mantle

Enriched mantle components

Several enriched components have been identified from the geochemistry of ocean island basalts (OIBs):

• EM1 (Enriched Mantle 1): possibly derived from recycled lower continental crust or pelagic sediments; characterized by moderately elevated $^{87}Sr/^{86}Sr$ and low $^{143}Nd/^{144}Nd$
• EM2 (Enriched Mantle 2): may represent recycled upper continental crust or terrigenous sediments; shows high $^{87}Sr/^{86}Sr$
• HIMU (High $\mu$, where $\mu = ^{238}U/^{204}Pb$): characterized by very radiogenic Pb isotopes, likely sourced from recycled oceanic crust that lost Pb during subduction dehydration

These components demonstrate that subducted material can survive in the mantle for billions of years and eventually be sampled by plume-derived volcanism.

Composition vs other planets

Comparing Earth's composition with other terrestrial bodies helps constrain what makes Earth unique and illuminates the processes that operated across the solar system during planet formation.

Terrestrial planets comparison

• Mercury: anomalously iron-rich (large core relative to its size), possibly due to a giant impact stripping its silicate mantle or preferential condensation of metal-rich material close to the Sun
• Venus: similar size and bulk density to Earth, but with a thick $CO_2$ atmosphere and no evidence of plate tectonics
• Mars: smaller and less dense, with a proportionally larger core sulfur content and a thin atmosphere; lacks active plate tectonics
• Earth is unique in having abundant surface liquid water, an active plate tectonic system, and a strong magnetic field generated by its liquid outer core
• Differences in volatile content, oxidation state, and size all influence how each planet differentiated and evolved

Earth-Moon system similarities

• The Moon most likely formed from debris ejected during a giant impact between proto-Earth and a Mars-sized body (Theia)
• Isotopic compositions of Earth and Moon are nearly identical for O, Cr, Ti, and W, which strongly constrains impact models
• The Moon is more depleted in volatile elements than Earth, consistent with high-temperature formation from impact debris
• The lunar highlands crust is Al-rich (anorthositic), somewhat analogous to Earth's continental crust in being a product of early differentiation
• The Moon's small iron core (~2% of its mass vs. ~33% for Earth) constrains how much of the impactor's core was incorporated

Analytical techniques

No single method can reveal Earth's full composition. Instead, geochemists combine direct sampling, remote sensing, and experimental petrology to build a coherent picture.

Seismic studies

• Seismic waves (P-waves and S-waves) travel at velocities that depend on the density, composition, and physical state of the material they pass through
• S-waves cannot propagate through liquids, which is how the liquid outer core was discovered
• Seismic tomography uses travel-time variations from many earthquakes to construct 3D images of mantle velocity structure, revealing convection patterns
• Major seismic discontinuities mark compositional or phase boundaries:
  • Moho (~7-70 km): crust-mantle boundary
  • 410 km: olivine → wadsleyite phase transition
  • 660 km: ringwoodite → bridgmanite + ferropericlase transition
• Free oscillations (normal modes) of the whole Earth constrain bulk properties like average density and elastic moduli

Geochemical sampling methods

• Crustal rocks: collected directly through fieldwork and drilling (deepest borehole: Kola Superdeep, ~12.3 km)
• Mantle xenoliths: fragments of mantle rock carried to the surface by deep-sourced volcanic eruptions (especially kimberlites), providing direct samples of the upper mantle
• MORBs: sample the composition of the depleted upper mantle source
• OIBs: may tap deeper, less depleted mantle reservoirs
• High-precision analytical instruments: ICP-MS (inductively coupled plasma mass spectrometry), TIMS (thermal ionization mass spectrometry), and SIMS (secondary ion mass spectrometry) measure elemental and isotopic compositions at increasingly fine spatial and concentration scales

Implications for geodynamics

Earth's composition doesn't just describe what the planet is made of; it controls how the planet behaves. Density contrasts, viscosity variations, and heat production all stem from compositional differences.

Mantle convection

• Compositional (chemical) density differences combine with thermal density differences to drive convection
• Phase transitions at 410 km and 660 km can either enhance or impede convective flow depending on the sign of their Clapeyron slopes
• Viscosity varies with composition, temperature, water content, and grain size, all of which influence convection vigor and style
• Melt extraction at mid-ocean ridges creates a compositionally depleted, buoyant residue that forms the lithospheric mantle
• Large Low Shear Velocity Provinces (LLSVPs) at the base of the mantle are compositionally distinct regions that may represent primordial material, accumulated subducted slabs, or both

Plate tectonics drivers

• Slab pull: the negative buoyancy of cold, dense subducting oceanic lithosphere is the dominant force driving plate motions
• Ridge push: the elevated topography of mid-ocean ridges (caused by hot, less dense mantle) creates a gravitational push on the plate
• Mantle drag: viscous coupling between convecting mantle and the base of plates can either drive or resist plate motion
• Continental buoyancy: continental lithosphere is compositionally buoyant (lower Fe/Mg, higher Si/Mg than oceanic lithosphere), which is why continents resist subduction and persist at the surface
• Slab dehydration: water released from subducting slabs lowers the solidus of the overlying mantle wedge, triggering partial melting and arc volcanism