Stellar Evolution
The Third Dredge-Up: How AGB Stars Bring Carbon and s-Process Elements to the Surface
Every few tens of thousands of years, a dying Sun-like star burps helium in a runaway flash so violent it powers up to 10⁸ L☉ for a few hundred years — and in the aftermath, the star's own outer convection zone plunges inward and hauls freshly minted carbon and heavy s-process elements up to the surface. This recurring mixing event is the third dredge-up (TDU or 3DU), the process by which thermally pulsing asymptotic-giant-branch (TP-AGB) stars turn oxygen-rich into carbon-rich and seed the galaxy with roughly half of all elements heavier than iron built by slow neutron capture.
It is the physical reason carbon stars, S-type stars, and technetium-line giants exist — and the main channel by which low- and intermediate-mass stars (~1.5–8 M☉) chemically enrich the interstellar medium.
- RegimeThermally pulsing AGB, cores ~0.6–1.0 M☉
- Occurs inStars ~1.5–8 M☉ (below ~1.5 M☉ no TDU)
- Driven byConvective envelope penetration after a He-shell flash (thermal pulse)
- Key numberEfficiency λ = ΔM_dredge / ΔM_core, ~0 to ~1
- First describedIben, Truran, Sackmann 1970s; Tc found by Merrill 1952
- Observed withOptical/near-IR spectroscopy — C₂, CN, ZrO bands, ⁹⁹Tc lines, Rb/Zr/Ba/La/Pb
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
What it is and why it matters
The third dredge-up is a recurrent mixing episode on the thermally pulsing asymptotic giant branch, the final nuclear-burning stage of stars born with roughly 1–8 M☉. By this point the star has a degenerate carbon–oxygen core, surrounded by a thin helium-burning shell and a hydrogen-burning shell, all wrapped in a vast, tenuous convective envelope. Energy is produced unstably: the He shell flashes every 10⁴–10⁵ years in a thermal pulse, and after each flash the convective envelope reaches down into layers that were just cooked by helium and neutron-capture reactions.
The consequence is chemical transformation. Freshly synthesized ¹²C and heavy elements built by the slow neutron-capture (s-) process — strontium, zirconium, barium, lanthanum, lead — are carried to the surface. Over many pulses the surface carbon-to-oxygen ratio can climb past 1, converting an ordinary M-type giant into an S-type and then a carbon star. AGB stars processed this way, then stripped by strong winds, are the dominant galactic source of carbon and about half of the s-process nuclei.
The mechanism, step by step
The cycle repeats on the TP-AGB. (1) Interpulse: the H-burning shell quietly adds helium ash to the intershell region, the mass gap between the H and He shells. (2) Thermal pulse: when enough He accumulates, it ignites degenerately in a helium-shell flash — a runaway that briefly reaches luminosities up to ~10⁸ L☉ and drives a convective zone through the intershell, dredging ³α-processed ¹²C (and a little ¹⁶O from ¹²C(α,γ)¹⁶O) outward within the He region.
(3) Power-down and dredge-up: the flash quenches, the intershell expands and cools, the H shell is temporarily extinguished, and the outer convective envelope sweeps inward past the (now inactive) H-shell boundary into the top of the intershell. This is the third dredge-up proper: carbon- and s-process-rich material is mixed into the envelope and appears at the surface. (4) The ¹³C pocket: as the envelope retreats, a few protons diffuse into the carbon-rich intershell; ¹²C(p,γ)¹³N(β⁺)¹³C makes a thin ¹³C pocket, which during the next interpulse releases neutrons via ¹³C(α,n)¹⁶O — the main s-process neutron source that builds the heavy elements later dredged up.
Characteristic numbers, scales, and the efficiency criterion
The bookkeeping quantity is the dredge-up efficiency parameter λ = ΔM_dredge / ΔM_core, the mass mixed to the surface divided by the core-mass growth during the preceding interpulse. λ = 0 means no dredge-up; λ ≈ 1 means the envelope claws back everything the H shell added. Models find λ growing from ~0 toward ~0.3–0.9 as the star evolves, with a minimum core mass of about 0.58–0.62 M☉ (roughly stars ≥ 1.5 M☉) needed for TDU to switch on at all.
Typical scales: thermal pulses recur every ~10⁴–10⁵ yr (shorter for more massive cores); each flash lasts a few hundred years; intershell masses are ~10⁻²–10⁻³ M☉. The neutron exposure is modest — the s-process needs ~0.1–1 neutron per iron seed. Two neutron sources compete: ¹³C(α,n)¹⁶O, active at ~9 × 10⁷ K during interpulse in low-mass stars, and ²²Ne(α,n)²⁵Mg, which switches on only above ~3 × 10⁸ K inside the pulses of intermediate-mass stars and yields a higher neutron density (favoring Rb over Sr/Zr).
How it is observed and detected
The smoking gun is technetium. All Tc isotopes are radioactive; ⁹⁹Tc, made by the s-process, has a half-life of only 2.13 × 10⁵ yr — comparable to the interpulse time. Paul Merrill's 1952 detection of Tc absorption lines (notably near 4238, 4262, 4297 Å) in S-type giants was decisive proof that heavy-element synthesis happens inside stars and reaches the surface, since the element cannot be primordial. Tc-rich giants are direct evidence of recent third dredge-up.
More broadly, TDU is read from optical and near-IR spectroscopy: molecular bands of C₂, CN and CH (carbon stars), ZrO and LaO (S stars), and atomic lines of Sr, Y, Zr, Ba, La, Ce and Pb quantify s-process enrichment; the [hs/ls] and [Pb/hs] ratios trace neutron exposure and metallicity. The ¹²C/¹³C ratio and C/O ratio track the carbon injection. Surveys and instruments — HERMES/GALAH, APOGEE, Gaia, and high-resolution echelle spectrographs (UVES, HERMES-Mercator) — map these signatures across thousands of evolved stars, and the enrichment persists into the ejected planetary nebulae.
Where it operates and how it differs from related effects
Third dredge-up is confined to the TP-AGB phase of stars from about 1.5 to 8 M☉; below ~1.5 M☉ the envelope never penetrates efficiently, and above ~8 M☉ stars ignite carbon and evolve toward core collapse instead. It is distinct from the first dredge-up (base of the RGB, mixing CN-cycle products) and the second dredge-up (early AGB in stars ≳ 4 M☉, bringing up He and N). Only the third dredge-up delivers freshly made ¹²C and s-process nuclei, and only it is recurrent.
A crucial competitor in the intermediate-mass range (≳ 4–5 M☉) is hot-bottom burning: the convective envelope's base gets hot enough (≳ 6 × 10⁷ K) to run CNO cycling, converting the dredged-up ¹²C into ¹⁴N and even making ⁷Li. This can prevent C/O from exceeding 1, so the most massive AGB stars often stay oxygen-rich despite active TDU. It also depends sensitively on convective overshoot at the envelope boundary, which sets both λ and the size of the ¹³C pocket.
Open questions and significance
The third dredge-up sits at the heart of the largest uncertainties in low-mass stellar nucleosynthesis. The efficiency λ and the onset core mass vary substantially between codes because they hinge on the treatment of convective boundary mixing / overshoot, which no first-principles theory fixes. The formation of the ¹³C pocket — how many protons penetrate, over what mass — is likewise parametrized, yet it controls the entire s-process yield; magnetic buoyancy, rotation, gravity waves, and opacity-driven mixing are all proposed drivers and none is settled.
These questions matter well beyond stellar interiors. AGB yields calibrated on TDU set the carbon, nitrogen, fluorine and s-process budgets in galactic chemical-evolution models, explain carbon-enhanced metal-poor (CEMP-s) stars polluted by an AGB companion, and account for anomalous isotopes in presolar SiC grains recovered from meteorites — direct condensates of AGB winds. Pinning down third dredge-up is thus essential to reading the chemical fossil record of the Milky Way.
| Episode | Evolutionary stage | What is mixed up | Composition signature |
|---|---|---|---|
| First dredge-up | Base of the red-giant branch (RGB) | CN-cycle products from H burning | ¹²C/¹³C drops to ~20–30; ¹⁴N up; ³He, Li changes |
| Second dredge-up | Early AGB, stars ≳ 4 M☉ | He- and N-rich layers after core He exhaustion | Surface He and ¹⁴N enhanced; ¹²C, ¹⁶O reduced |
| Third dredge-up (3DU/TDU) | Thermally pulsing AGB, recurrent | ¹²C (triple-α) + s-process elements from the intershell | C/O rises past 1 → carbon star; Tc, Zr, Ba, Pb up |
| Hot-bottom burning (competes) | TP-AGB, stars ≳ 4–5 M☉ | CNO processing at the envelope base (not mixing up) | Converts ¹²C→¹⁴N, makes ⁷Li; can suppress carbon-star formation |
Frequently asked questions
Why is it called the 'third' dredge-up?
It is the third distinct episode in which a star's convective envelope reaches down and mixes nuclear-processed material to the surface. The first dredge-up happens at the base of the red-giant branch (bringing up CN-cycle products), the second on the early AGB for stars above ~4 M☉ (bringing up helium and nitrogen), and the third recurs during the thermally pulsing AGB, uniquely delivering fresh carbon and s-process elements.
What actually drives the mixing?
A helium-shell flash (thermal pulse). Helium ignites unstably in a thin shell, briefly spiking the luminosity and driving a convective zone through the carbon-rich intershell. After the flash subsides the intershell cools and the hydrogen shell shuts off, allowing the star's outer convective envelope to sweep inward past the old hydrogen-burning boundary and carry the enriched material up to the surface.
How does the third dredge-up make a carbon star?
Each dredge-up injects triple-alpha carbon (¹²C) into the envelope while barely changing the oxygen. After enough pulses the surface carbon-to-oxygen number ratio, C/O, climbs above 1. Because CO locks up whichever of C or O is scarcer, an envelope with C/O > 1 has free carbon that forms C₂, CN and CH — the defining molecular bands of a carbon (C-type) star. Intermediate stages with C/O near 1 appear as S-type stars.
Where do the s-process elements come from during this?
They are built in the intershell by slow neutron capture. Between pulses, a thin ¹³C pocket releases neutrons through ¹³C(α,n)¹⁶O; in more massive AGB stars the hotter pulses also fire ²²Ne(α,n)²⁵Mg. Those neutrons are captured by iron-peak seeds, walking them up to strontium, barium, lead and beyond. The third dredge-up then mixes these newly made heavy elements to the surface.
Why is technetium such important evidence?
Technetium has no stable isotopes; the s-process isotope ⁹⁹Tc has a half-life of only 2.13 × 10⁵ years, similar to the interval between thermal pulses. If it appears in a stellar spectrum, it must have been made recently and dredged up on a comparable timescale — it cannot be inherited from the star's birth cloud. Paul Merrill's 1952 detection of Tc lines in S stars proved that heavy elements are synthesized inside stars.
What is the dredge-up efficiency parameter λ?
λ is defined as the mass mixed to the surface by one dredge-up divided by the mass the hydrogen shell added to the core during the preceding interpulse: λ = ΔM_dredge / ΔM_core. λ = 0 means no dredge-up; λ ≈ 1 means the envelope recovers essentially all the freshly burned material. It typically rises from near zero to a few tenths as the star evolves, but its exact value is one of the biggest uncertainties in AGB modeling because it depends on how convective overshoot is treated.