Conservative vs Non-Conservative Mass Transfer in Binaries

What it is: the physical basis

Stars in a close binary are each surrounded by a teardrop-shaped Roche lobe, the region within which their gravity dominates. When a star expands — because it evolves off the main sequence, or because the orbit shrinks — it can fill and overflow that lobe. Gas then streams through the inner Lagrange point (L1) and falls toward the companion. This is Roche-lobe overflow (RLOF), the engine of binary mass transfer.

The central question is what happens to that gas once it reaches the accretor. In the conservative limit, the accretor swallows all of it: the system's total mass M = M_d + M_a and its orbital angular momentum J are both unchanged. In the non-conservative case, the accretor cannot retain everything — it may already be spinning near break-up, or radiating near its Eddington limit — so a fraction of the material is driven back out of the system.

• β ≡ |Ṁ_a / Ṁ_d| is the accretion efficiency.
• β = 1 means fully conservative; β = 0 means every gram donated is re-ejected.

The mechanism: efficiency, angular momentum, and orbital response

Mass transfer reshapes the orbit because moving mass between the stars moves angular momentum between them. For conservative transfer, J and M are fixed, and the orbital separation obeys a = J²/(G M_a² M_d²) × (M/…), which simplifies to the well-known scaling a·M_a²·M_d² = constant. The consequence is dramatic: the separation reaches a minimum when the two masses are equal, then the orbit widens once mass flows from the lighter to the heavier star.

For non-conservative transfer, the lost mass carries away angular momentum, and the outcome depends on where it leaves. The standard prescription is isotropic re-emission: matter is accreted, then blown off from the vicinity of the accretor (like a fast wind or a jet), carrying that star's specific orbital angular momentum. In accretor units this gives a specific angular momentum γ_iso = q⁻¹ where q = M_d/M_a. Two other channels are the Jeans mode (fast wind from the donor) and mass lost through the outer L2 point, which drains far more angular momentum and can shrink the orbit.

• Widening or shrinking hinges on β and the ejection site.
• Isotropic re-emission generally lets the orbit widen while removing modest angular momentum.

Key quantities and a worked example

Consider a massive binary with a donor of M_d = 15 M_☉ and accretor M_a = 10 M_☉ (mass ratio q = 1.5, period a few days). The donor finishes core hydrogen burning, expands, and fills its Roche lobe. Because it is the more massive star, transfer runs on its thermal (Kelvin–Helmholtz) timescale, τ_KH ≈ G M²/(R L) ≈ 10⁴–10⁵ years, giving rates of roughly 10⁻³–10⁻⁴ M_☉/yr.

If the accretor is already near critical rotation or its Eddington limit L_Edd ≈ 3.2×10⁴ (M_a/M_☉) L_☉ ≈ 3×10⁵ L_☉, it cannot absorb such a torrent. Suppose β = 0.2: only 0.2 M_☉ is retained for every 1 M_☉ donated, and 0.8 M_☉ escapes via isotropic re-emission. As the mass ratio drops through unity, the separation stops shrinking and begins to grow.

• Thermal-timescale (fast) phase: 10⁴–10⁵ yr, high Ṁ.
• Nuclear-timescale (slow) phase: 10⁷–10⁸ yr, low Ṁ (~10⁻⁷ M_☉/yr) — the long-lived semi-detached state we actually observe.
• Eddington-limited accretion caps retention onto compact accretors, forcing low β.

How it's observed and where it appears

We infer mass-transfer efficiency indirectly, by matching observed binaries to evolutionary models. Algol-type semi-detached binaries are the archetype: the presently less-massive star fills its Roche lobe and is the more evolved one — direct evidence that mass has already been swapped. Their period changes, measured over decades from eclipse timing (O−C diagrams), constrain how much mass and angular momentum leave the system.

• X-ray binaries (e.g., SS 433, Cygnus X-1, ultraluminous X-ray sources) show non-conservative transfer directly: jets and disk winds carry away material the compact object cannot accrete.
• Be/X-ray binaries in the Small Magellanic Cloud have been used as population-level tracers of β, favoring low-to-moderate efficiency.
• Double-lined eclipsing binaries yield precise masses and radii, anchoring the models.

Chemically, the accretor can be enriched in CNO-processed or helium-rich gas dredged from the donor's interior — a fingerprint of past transfer visible in stellar spectroscopy.

Comparison to related regimes and cousins

Mass transfer is classified by when the donor overflows, following Kippenhahn & Weigert:

• Case A — during core hydrogen burning (main sequence); slow, often long-lived.
• Case B — after core hydrogen exhaustion, before helium ignition (the most common channel for massive binaries).
• Case C — after core helium burning, from an evolved supergiant.

Conservative vs non-conservative is an orthogonal axis to this A/B/C classification — any case can be more or less efficient. It also differs from common-envelope evolution, the dramatic alternative that occurs when transfer is dynamically unstable: the donor's envelope engulfs both stars and is ejected, shrinking the orbit by factors of hundreds. Stable non-conservative RLOF and unstable common-envelope inspiral are the two great forks in binary evolution; which one occurs depends on the mass ratio, the donor's structure (radiative vs convective envelope), and β itself.

Significance, famous cases, and open questions

The Algol paradox — why does the less massive star in β Persei look older? — was the puzzle that revealed mass transfer. Its resolution (Crawford 1955; Kopal, Morton) was that Algol's donor was originally the heavier, faster-evolving star and has since transferred most of its mass, inverting the ratio. This single insight launched the entire field of binary stellar evolution.

Today, β is one of the largest sources of uncertainty in modeling gravitational-wave progenitors. Whether two massive stars become a merging black-hole or neutron-star binary depends sensitively on how much mass and angular momentum leave during each transfer episode. Recent 2026 work even challenges the textbook assumption that isotropic re-emission always stabilizes transfer — for radiative donors it can hasten instability, tipping systems toward common envelope.

• Open: the true value and variation of β across mass, metallicity, and orbital period.
• Open: how accretors spin up and whether that self-regulates the flow.
• Open: the exact boundary between stable transfer and dynamical instability.

Conservative vs non-conservative mass transfer: how the key physics differs

Property	Conservative (β = 1)	Non-conservative (β < 1)
Total system mass	Conserved (Ṁ_a = −Ṁ_d)	Decreases; fraction (1−β) is lost
Orbital angular momentum	Conserved	Lost with escaping matter
Orbit response after mass-ratio reversal	Widens (a increases)	Widens, often faster; can also shrink if AM loss is large
Common ejection mode	None (all accreted)	Isotropic re-emission, γ ≈ q⁻¹ (accretor units)
Accretor spin-up	Modest	Can spin to critical rotation, triggering ejection
Typical β for massive binaries	Idealized limit	≈ 0.1–0.5 (very uncertain)