Planetary growth compresses an enormous amount of change into a deceptively simple verb: accrete. Earth accumulated matter, but it did not do so at a constant rate or by one collision mechanism. Dust, pebbles, planetesimals, planetary embryos, and late impactors participated at different stages. Depending on what milestone is chosen—first seed, most of the mass, Moon-forming collision, or final minor additions—Earth’s construction has a different finishing date.
The First Size Barrier
The inner protoplanetary disk contained countless solid particles orbiting through gas. Very small grains could adhere after low-speed contact. Larger aggregates, however, tended to bounce, fragment, or drift sunward before gradual sticking could turn them into planets. Modern formation models therefore investigate shortcuts across this size barrier. Turbulence and aerodynamic effects can concentrate particles; a dense swarm can then collapse under its own gravity and produce planetesimals tens or hundreds of kilometers across.
Once an object reached planetesimal scale, its gravity increased the area from which it could collect material. A collision partner did not need to be aimed directly at its surface, because gravitational focusing could bend the smaller body’s path inward. This feedback encouraged runaway growth: the largest objects grew faster than their neighbors.
From Planetesimals to Embryos
Runaway growth eventually shifted into a more regulated stage. Large bodies stirred nearby planetesimals into faster, more inclined orbits, reducing the efficiency of capture. A set of separated planetary embryos emerged, each dominating a feeding zone. Within the rocky planet region, these embryos may have approached the mass of the Moon or Mars while gas still surrounded the young Sun.
The relative importance of planetesimals and small drifting “pebbles” remains debated. Pebble accretion can be highly efficient under the right disk conditions, but isotopic differences among meteorites and the architecture of the inner Solar System constrain how freely material could cross the disk. Earth’s bulk composition suggests that most of its mass came from inner Solar System material, with smaller contributions from chemically distinct sources.
Giant Impacts Finished the Main Assembly
After the gas disk dispersed, aerodynamic damping weakened and embryos occupied unstable, crossing orbits. Numerical simulations show a prolonged phase of giant impacts. These were not perfectly inelastic collisions in which every gram stayed on the largest body. An oblique encounter might graze and escape; a powerful hit could eject a mantle; debris could remain in orbit, reaccrete, or be lost. Growth was therefore a balance sheet of additions and removals.
Earth probably acquired roughly 99 percent of its final mass within about 60 to 100 million years of the Solar System’s first solids, although estimates depend on the model and isotope system used. Some chronologies favor faster acquisition of most material, followed by a late giant collision. The Moon-forming impact provides a useful marker near the end of major accretion, but its precise age is not settled by a single measurement.
Radioactive Nuclides Act as Formation Clocks
Scientists cannot date accretion by finding one intact “first Earth” rock. Instead, they study how elements were redistributed as the planet grew and formed a core. The hafnium–tungsten system is especially valuable. Radioactive hafnium-182 decays to tungsten-182 with a half-life of roughly nine million years. Hafnium prefers silicate, whereas tungsten is drawn into metal. When metal separated into the core, it changed the parent-to-daughter relationship left in the mantle.
Because hafnium-182 disappeared early in Solar System history, present tungsten isotope ratios record when this separation occurred. The interpretation is not a simple stopwatch reading: accretion delivered new metal repeatedly, impacts mixed reservoirs to different degrees, and assumptions about the impactors affect calculated dates. Even so, short-lived isotope systems demonstrate that core formation and planetary accumulation began promptly and unfolded over tens of millions of years.
Other clocks and chemical tracers add constraints. Uranium–lead ages date the Solar System’s earliest solids and later crystallization events. Highly metal-loving elements in the mantle reveal a small “late veneer” added after most core separation. Isotopic similarities and differences among Earth, the Moon, Mars, and meteorites narrow the possible sources and timing of the material.
Accretion Never Had a Sharp Final Collision
Earth continued sweeping up leftover bodies after the principal giant-impact phase. The mass involved was small compared with the finished planet, but it mattered chemically. Late arrivals may have supplied some volatile compounds and restored metal-loving elements to the mantle. Much later impacts are part of Earth’s ongoing collision history, not evidence that the planet was still assembling in the same fundamental sense.
Saying that accretion built Earth “in millions of years” is therefore accurate but incomplete without scale. Planetesimals appeared quickly; embryos followed; most of Earth accumulated over tens of millions of years; and minor additions continued after the Moon-forming event. The process was fast compared with Earth’s 4.54-billion-year lifetime, yet long enough for repeated melting, core formation, orbital rearrangement, and collisions between entire young worlds.