How Stars Form

Overview

Stars form when the densest parts of cold interstellar clouds become unable to support their own weight and collapse. The physics is a contest between gravity pulling inward and thermal pressure, turbulence, magnetic fields and radiation pushing back — and the outcome, repeated across a galaxy, sets how many stars of each mass exist.

• Star formation happens inside molecular clouds at temperatures around 10–20 K, cold enough for hydrogen to be molecular rather than atomic.
• Collapse is not a single event: a cloud fragments, so stars are usually born in groups rather than one at a time.
• Conservation of angular momentum forces the infalling gas into a disc — which is why planet formation is a by-product of star formation, not a separate process.
• A protostar becomes a star when its core reaches roughly 10 million kelvin and hydrogen fusion begins supplying the energy it radiates.

Molecular clouds: the raw material

The interstellar medium is not uniform. Its coldest, densest phase is organised into molecular clouds — regions where hydrogen exists as H₂ rather than as individual atoms, shielded from starlight by dust. Giant molecular clouds reach tens of parsecs across and can contain hundreds of thousands of solar masses of gas at temperatures near 10 kelvin.

H₂ itself is almost invisible at these temperatures, so astronomers map clouds using tracer molecules. Carbon monoxide is the workhorse: it is abundant, and its rotational transitions radiate at millimetre wavelengths that penetrate dust. Dust emission at far-infrared and submillimetre wavelengths, mapped by facilities such as Herschel and ALMA, gives an independent view of the same structures.

Dense cores and the balance that fails

Within a cloud, turbulence and magnetic fields carve out filaments, and along those filaments gas accumulates into dense cores roughly 0.1 parsec across. A core is stable while its internal pressure — thermal, turbulent and magnetic — can hold up the weight of the gas above it.

The classical statement of when that fails is the Jeans criterion: for a given temperature and density there is a mass above which gravity wins. Lower temperature and higher density both lower that threshold, which is why collapse begins in the coldest, densest material. Real cores are more complicated than the idealised calculation — magnetic support and turbulence both matter — but the direction of the argument holds.

Fragmentation: why stars are born in groups

As a collapsing region contracts, its density rises much faster than its temperature while the gas can still radiate away the heat of compression. That drives the local Jeans mass down, so sub-regions inside the collapsing cloud become independently unstable and begin collapsing on their own. The cloud fragments.

This is why the overwhelming majority of stars form in clusters and associations rather than in isolation, and why binary and multiple systems are common. It is also the origin of a long-standing question in the field: the distribution of stellar masses that emerges — the initial mass function — is remarkably similar in very different environments, and explaining that regularity from first principles remains an active problem.

The protostar and its disc

Once the centre of a fragment becomes dense enough to trap its own radiation, it stops cooling efficiently, heats up, and forms a pressure-supported object: a protostar. It is still deeply embedded in the infalling envelope and is visible only at infrared and longer wavelengths, where the surrounding dust is transparent.

The infalling gas carries angular momentum. It cannot fall straight in, so it settles into a rotating circumstellar disc, and material reaches the star by working its way inward through that disc. Observationally this sequence is classified by the shape of the spectral energy distribution — the Class 0, I, II and III scheme — which tracks how much envelope is left relative to the star and disc.

ALMA's 2014 image of HL Tauri showed such a disc with concentric gaps at an age of only about a million years, direct evidence that discs are structured — and plausibly already forming planets — very early.

Jets, outflows, and how a star sheds angular momentum

Accretion is not tidy. Young stellar objects drive fast, narrow, bipolar jets along their rotation axes, together with wider slower outflows. Where those jets ram into surrounding cloud material they excite bright shock-heated knots — Herbig–Haro objects — which are among the most direct visual signatures of ongoing star formation.

Outflows matter physically, not just aesthetically. They remove angular momentum that the collapsing gas would otherwise have to keep, and they inject energy and momentum back into the parent cloud, disrupting it and helping to limit how much of the cloud ever becomes stars.

Pre-main-sequence evolution

When the envelope clears, the object becomes optically visible as a pre-main-sequence star: a T Tauri star at low masses, or a Herbig Ae/Be star at intermediate masses. It is larger and more luminous than it will be as a mature star, and it is still shrinking. Its energy comes from gravitational contraction, not yet from hydrogen fusion.

Deuterium fusion ignites first, at a lower temperature than ordinary hydrogen fusion, and briefly slows contraction. On a Hertzsprung–Russell diagram low-mass stars descend a nearly vertical convective track before turning onto a nearly horizontal radiative one as the interior becomes stable against convection.

Ignition and arrival on the main sequence

Contraction stops when the core reaches roughly 10 million kelvin and hydrogen fusion becomes self-sustaining. The star then sits on the zero-age main sequence, in stable equilibrium: fusion supplies exactly the energy it radiates, and the resulting pressure balances gravity. This is the longest and most stable phase of a star's life.

Mass controls almost everything about the journey and the destination. A star of about a solar mass takes on the order of tens of millions of years to reach the main sequence; a massive star gets there in a few hundred thousand years and begins ionising its birth cloud while still accreting. Below roughly 0.08 solar masses an object never reaches core temperatures sufficient for sustained hydrogen fusion at all: it becomes a brown dwarf, which fuses deuterium briefly and then simply cools.

How we know: the observational evidence

• Infrared and submillimetre surveys detect protostars still buried in their envelopes, invisible at optical wavelengths — the basis of the Class 0/I/II/III sequence.
• Resolved discs imaged by ALMA and by JWST show the rotating structures the theory requires, including gaps and rings.
• Herbig–Haro objects and molecular outflows trace jets actively driven by accreting young stars.
• Young clusters such as the Orion Nebula Cluster contain stars of many masses at essentially the same age, matching the prediction that clouds fragment rather than forming single stars.
• Statistical surveys of clusters recover a similar initial mass function across widely different environments, constraining any successful theory.

What is still open

Several parts of this picture are genuinely unsettled, and it is worth stating that plainly rather than smoothing it over. How the most massive stars form is debated: their radiation pressure is strong enough to oppose further accretion, and competing models — monolithic collapse of a very massive core versus competitive accretion within a cluster — make different predictions that observations have not yet decisively separated.

The relative importance of magnetic fields versus turbulence in supporting clouds remains an active question, as does the physical origin of the initial mass function and how efficiently a given cloud converts its gas into stars. Disc fragmentation as a route to forming companions and giant planets is likewise still being worked out.

Continue in the data

Catalogues, hubs, and reference pages that hold the underlying records for this topic.

• Stellar astrophysicsThe physics of stellar interiors, evolution, and endpoints.
• Deep sky catalogueNebulae, clusters, and star-forming regions with catalogue data.
• StarsThousands of catalogued stars with measured parameters and provenance.
• AstrochemistryThe molecules that trace and cool collapsing clouds.

Frequently asked

Where do stars form?

Inside molecular clouds — the coldest, densest phase of interstellar gas, where hydrogen is molecular and dust shields the interior from starlight. In the Milky Way these clouds are concentrated in the spiral arms and the galactic disc. Nearby examples visible from Earth include the Orion Nebula complex, about 1,300 light-years away.

What causes a molecular cloud to collapse?

Collapse begins where gravity overcomes the internal pressure supporting the gas. Because that threshold falls as temperature drops and density rises, the coldest and densest cores go first. External triggers can help — a nearby supernova shock, spiral-arm compression, or collision between clouds — but a sufficiently dense core will become unstable without one.

What is a protostar?

A protostar is the central object that forms once a collapsing fragment becomes dense enough to trap its own radiation and stop cooling freely. It is supported by pressure and is already luminous, but its energy comes from gravitational contraction and accretion rather than from hydrogen fusion. It is typically still buried in an infalling envelope and detectable mainly at infrared wavelengths.

When does a protostar become a star?

When its core reaches roughly 10 million kelvin and hydrogen fusion becomes self-sustaining, supplying the energy the object radiates. At that point contraction halts and the star settles onto the zero-age main sequence. Objects below about 0.08 solar masses never reach that threshold and become brown dwarfs instead.

Why do young stars have discs?

Because the collapsing gas is rotating. Angular momentum is conserved, so material cannot fall directly onto the central object; it settles into a rotating disc and spirals inward from there. That disc is also the raw material for planets, which is why planet formation is a natural consequence of star formation rather than a separate event.

How long does star formation take?

It depends strongly on mass. The deeply embedded protostellar phase lasts on the order of a hundred thousand years. A star of roughly one solar mass then takes tens of millions of years of contraction to reach the main sequence, while a massive star can complete the whole sequence in a few hundred thousand years — fast enough that it starts ionising its birth cloud before accretion has finished.

Can astronomers actually watch stars being born?

Not a single star from start to finish — the process is far longer than human timescales. But because star-forming regions contain many objects at different stages simultaneously, the sequence can be assembled observationally: infrared surveys catch embedded protostars, ALMA resolves their discs, and Herbig–Haro objects show jets in action right now.