🛸 Orbital Mechanics · Space
📅 Apr 2026 ⏱ ~9 min read 🟢 All levels

Lagrange Points — L1–L5, JWST, SOHO and Trojan Asteroids

In a system of two large masses — say, the Sun and Earth — there are five special positions where a small third object can orbit in perfect synchrony with them, maintaining its position relative to both. These are the Lagrange points: parking spots used by some of our most important space telescopes, and naturally claimed by thousands of asteroids.

The Restricted Three-Body Problem

The general three-body problem — three bodies interacting through gravity — has no closed-form analytical solution in general. Henri Poincaré proved in 1890 that its solutions are chaotic: tiny changes in initial conditions lead to wildly different outcomes.

The restricted three-body problem (R3BP) simplifies this by assuming the third body has negligible mass and thus does not affect the orbits of the two primary bodies. In this setting, Joseph-Louis Lagrange found in 1772 that there are exactly five equilibrium points where the combined gravitational and centrifugal forces on the small body exactly balance — so it remains stationary in the rotating reference frame.

In the co-rotating frame (rotating with the two primary bodies), the effective potential Φeff includes both gravitational terms and a centrifugal term:

Effective potential (co-rotating frame):
Φeff(x,y) = −GM1/r1 − GM2/r2 − ½ω²(x²+y²)

r1, r2 = distances from the third body to M1 and M2
ω = orbital angular velocity of the primary system

The five Lagrange points are the saddle and extremal points of this potential (where ∇Φeff = 0).

The Jacobi constant CJ = −2Φeff − v² is conserved for the third body — it partitions the space into regions accessible or forbidden; the boundaries are called zero-velocity curves (or Hill curves), used to study which trajectories are gravitationally allowed.

The Five Lagrange Points

The geometry below describes the Sun–Earth system, but the same points exist for any two-body system (Earth–Moon, Jupiter–Sun, etc.).

L1
Between the two bodies

Located on the line connecting M1 and M2, between them. For Sun–Earth: ~1.5 million km sunward of Earth (about 1% of 1 AU). Provides an unobstructed view of the Sun. Unstable — requires stationkeeping. Hosts: SOHO, DSCOVR, L1 halo orbits.

L2
Beyond the smaller body

On the M1–M2 line, on the far side of M2 from M1. For Sun–Earth: ~1.5 million km anti-sunward of Earth. Permanently shaded from the Sun by Earth (with a solar shield). Unstable. Hosts: James Webb Space Telescope, Gaia, Planck, Herschel.

L3
Directly opposite

On the M1–M2 line, on the far side of M1 from M2 — directly opposite M2 as seen from M1. For Sun–Earth: just beyond the Sun. Always hidden from Earth. Unstable; practically unusable for spacecraft. Historically the location of sci-fi “Counter-Earth”.

L4
60° ahead of M2

Forms an equilateral triangle with M1 and M2, leading M2 by 60° in its orbit. Stable provided M1/M2 > 24.96. Jupiter’s L4 hosts thousands of Greek Trojan asteroids. Earth’s L4 hosts asteroid 2010 TK7.

L5
60° behind M2

Equilateral triangle with M1 and M2, trailing M2 by 60°. Same stability conditions as L4. Jupiter’s L5 hosts thousands of Trojan Trojans (Trojan group). The proposed space colony/manufacturing concept (O’Neill cylinders) commonly cited Earth’s L4 or L5.

1.50 Gm Distance from Earth to Sun–Earth L1 and L2 (≈ 1% of 1 AU)
~29.5 days JWST halo orbit period around L2
~12,000 Known Jupiter Trojan asteroids sharing its L4/L5 points
24.96 Minimum mass ratio M1/M2 for L4/L5 stability (Routh criterion)

Stability: Why L4 and L5 Work

A naive argument says L1–L3 should be unstable: they are saddle points of the effective potential — a ball placed exactly there would roll away under any perturbation. L4 and L5 are local maxima of the effective potential, which sounds even worse.

However, stability is not just about potential energy. What actually matters is the full equations of motion including the Coriolis force present in the rotating frame. Linearising the equations around L4/L5 gives a characteristic equation whose eigenvalues determine stability:

Routh stability criterion for L4/L5:
L4 and L5 are stable (eigenvalues purely imaginary ⇒ oscillation, not runaway) if and only if:

μ < μc = (1 − √(69)/9) / 2 ≈ 0.03852

where μ = M2/(M1+M2) is the mass parameter.

For Sun–Earth: μ ≈ 3×10−6 ≪ μcstable.
For Sun–Jupiter: μ ≈ 9.5×10−4 ≪ μcstable.
For Earth–Moon: μ ≈ 0.012 < μcstable.

Objects near L4/L5 undergo two superimposed oscillations: a fast epicyclic libration and a slow long-period libration (the tadpole and horseshoe orbits). Trojan asteroids follow tadpole orbits — elongated loops centred on L4 or L5 in the rotating frame, like a tadpole swimming around the Lagrange point.

L1, L2 and L3 are unstable: perturbations grow exponentially in time. A spacecraft placed there will drift away unless it uses onboard propulsion (stationkeeping) to return to the point periodically. The characteristic escape timescale is typically weeks to months, making stationkeeping manageable with small amounts of fuel.

Real Missions: JWST, SOHO, DSCOVR

James Webb Space Telescope (L2)

JWST launched on Christmas Day 2021 and arrived at the Sun–Earth L2 point 30 days later. L2 is ideal for an infrared telescope because:

JWST does not sit exactly at L2 but in a large halo orbit around it (roughly 800,000 km in amplitude), avoiding the shadow of Earth and Moon that would periodically block sunlight needed for its solar panels.

SOHO and DSCOVR (L1)

The Solar and Heliospheric Observatory (SOHO, 1995) and DSCOVR (2015) orbit around the Sun–Earth L1 point. L1 gives a continuous, unobstructed view of the Sun with about 1-hour advance warning of solar storms before they reach Earth — crucial for space weather forecasting and protecting satellites.

Other L2 Missions

L2 has become prime real estate for space observatories: Gaia (astrometry), Planck (CMB), Herschel (far-infrared and submillimetre). The proposed LISA (Laser Interferometer Space Antenna) gravitational wave detector will orbit the Sun in an Earth-trailing configuration related to L4/L5 geometry.

Trojan Asteroids

Jupiter’s gravitational dominance has captured thousands of asteroids into stable orbits near its L4 (Greek camp, ~60° ahead) and L5 (Trojan camp, ~60° behind) points. These Jupiter Trojans are primordial material from the early solar system — likely containing organic compounds and water ice, preserved in the outer solar system — and collectively may contain more material than the entire asteroid belt.

NASA’s Lucy mission (launched 2021) is on a 12-year journey to fly past seven Trojan asteroids, aiming to study this primordial material. It will be the first spacecraft to visit the Jupiter Trojans.

Earth also has a co-orbital companion: 2010 TK7, a ~300-m asteroid librating around the Sun–Earth L4 point. It follows a large, complex horseshoe–tadpole hybrid orbit and will remain co-orbital for millennia. Neptune has over 30 known Trojans; Mars has at least 9.

Horseshoe orbits: Some objects near L3–L4–L5 follow a wider figure-8–like horseshoe orbit in the rotating frame — approaching L4, travelling all the way around past L3 to L5, then back again over centuries. Cruithne (a near-Earth asteroid) follows such an orbit around the Sun–Earth system and is sometimes loosely called Earth’s “second moon”.

Hill Sphere and Sphere of Influence

The Hill sphere (also called Roche sphere) is the region around a body M2 within which it can hold satellites against the tidal force of the dominant body M1. Its radius is approximately:

Hill sphere radius:
rH ≈ a (M2 / 3M1)1/3

a = orbital semi-major axis of M2 around M1

Earth’s Hill sphere: rH ≈ 1.5 million km (the Moon at 384,000 km is well inside).
Jupiter’s Hill sphere: rH ≈ 53 million km.

The L1 and L2 points both lie approximately at the edge of the Hill sphere — which makes intuitive sense: they are the farthest positions where a third body can still be considered “associated” with M2 rather than being captured by M1.

A related concept is the sphere of influence (SOI), used in spacecraft trajectory planning: the radius within which a planet’s gravity dominates over the Sun’s for the purposes of patched-conic trajectory approximations. SOIEarth ≈ 924,000 km — slightly smaller than the Hill sphere.

Try It Yourself

Why L2 is 1.5 million km and not further: A naive guess might be that L2 is at the same distance as the Moon, or even further. But L1 and L2 are determined by the mass ratio of the two bodies, not by the orbital radius alone. The formula is approximately a × (MEarth/3MSun)1/3 ≈ 149.6 Gm × 0.01 = 1.5 Gm (1.5 million km).
🛸 Open Orbital Mechanics Simulation →