Gravitational Self-Energy and Extended Problems

Look at a cricket ball. Pick it up — 160 grams, 7 cm across, sitting quietly in your hand. Now ask a strange question: how much energy did it take to build that cricket ball? Not manufacture it — gravitationally build it. If the cork and leather were once a cloud of dust spread across all of space, how much work did gravity have to do pulling it together?

The answer for a 160 g sphere is a joke. A fraction of a nanojoule. You would never notice it. But ask the same question about the Earth — and the answer is 2.24 \times 10^{32} joules, which is roughly the total energy output of the Sun over three weeks, stored quietly in the fact that Earth is bound and not dust. Ask it about the Sun itself, and you get 2.3 \times 10^{41} joules — enough to power the Sun's luminosity for ten million years. For a neutron star, the gravitational binding energy is a few per cent of its entire rest-mass energy. The thing is made mostly of the energy of its own assembly.

This is gravitational self-energy. Together with a few other ideas — binary stars, tides, weightlessness, gravity assists — it closes out the gravitation chapter. All of it is built from the two tools you already have: Newton's F = GMm/r^2 and U = -GMm/r. No new physics, just richer applications.

Gravitational self-energy of a uniform sphere

Start with the basic question. You have a mass M spread out as dust all the way to infinity. You want to assemble it into a uniform solid sphere of radius R and uniform density \rho = M/\!\left(\tfrac{4}{3}\pi R^3\right). What is the total work done by gravity during this assembly? That negative of that work is the self-energy.

Assumptions. The sphere is built from a dust cloud of the same total mass M, with every dust particle starting at infinite separation from every other. The material is non-rotating, held together only by its own gravity, and has uniform density once assembled. We ignore pressure, rotation, and quantum-mechanical effects — all are small for ordinary astronomical bodies.

The "onion" construction

Imagine the sphere growing outward from a seed. At some intermediate stage it is a ball of radius r carrying mass

Why: mass equals density times volume, and the growing ball has volume \tfrac{4}{3}\pi r^3. The density \rho is fixed by the final state.

Now add the next thin shell: thickness dr, surface area 4\pi r^2, mass

Why: the volume of a thin spherical shell is surface area times thickness. Multiply by density to get its mass.

The energy of wrapping one shell

Bringing the shell of mass dm from infinity to wrap it onto a ball of mass m(r) is a problem you already solved: two separated masses, one concentric. Because the shell is being placed right at the surface of the already-assembled ball, its final separation from the ball's mass (which by the shell theorem acts as if concentrated at the centre) is r.

So the potential energy of that one shell, once placed, relative to its initial dispersed state at infinity, is

Why: this is the pair potential energy formula U = -GMm/r applied to the inner ball (mass m(r), treated as a point at the centre by the shell theorem) and the new shell (mass dm, at distance r). The shell's own self-energy is negligible because dm is infinitesimal and dm^2 is doubly so.

Substitute the expressions for m(r) and dm:

Why: multiply the two masses and divide by r. The r^3 \cdot r^2 / r gives r^4. All constants collect into the prefactor 16\pi^2 G \rho^2/3.

Integrate over all shells

The total self-energy is the sum of dU from the innermost shell (r = 0) to the last shell (r = R):

Why: \int_0^R r^4\,dr = R^5/5. Pull the constants through and evaluate.

Now rewrite \rho in terms of M. From M = \rho \cdot \tfrac{4}{3}\pi R^3, solve for \rho:

Substitute:

Why: the 16\pi^2 cancels. The R^5 / R^6 = 1/R. The fraction 9/15 = 3/5 after simplification. Every piece of apparatus from the integration dissolves, leaving one clean formula.

This is the self-energy result for a uniform solid sphere. It is the energy cost of existence for a gravitationally bound ball.

Reading the answer

• The minus sign. Same story as before: the assembled sphere is bound. It took gravity doing positive work to pull everything in; to disperse the sphere back to infinity, something external would have to do +3GM^2/(5R) of work — the gravitational binding energy.
• The M^2 scaling. Double the mass, and self-energy goes up by four. This is why small bodies (a cricket ball) have negligible self-energy, while big bodies (a star) are held together by immense gravitational binding.
• The 1/R scaling. Compress the sphere to half its radius and self-energy becomes twice as negative — more deeply bound. Neutron stars, with Earth-like mass compressed into 10-km spheres, are ferociously bound.
• The 3/5 coefficient. This is the signature of a uniform sphere. A centrally concentrated mass distribution (like a real star, which is denser at the core) has a more negative self-energy for the same M and R — the coefficient rises above 3/5.

A numerical feel

Plug in Earth: M = 5.97 \times 10^{24} kg, R = 6.37 \times 10^6 m.

The numerator is 3 \times 6.674 \times 35.64 \times 10^{37} = 713.7 \times 10^{37} (roughly); the denominator is 31.85 \times 10^6. So

The Sun radiates about 3.8 \times 10^{26} W. Dividing gives 2.24 \times 10^{32} / 3.8 \times 10^{26} \approx 5.9 \times 10^{5} s, or about a week. Earth's gravitational binding energy equals roughly a week of the Sun's total output. Dismantling the Earth — bit by bit, out to infinity — would take energy on that scale. This is not a hypothetical exercise; it sets the upper limit on how much energy any catastrophe (asteroid impact, gamma-ray burst) can deposit in the Earth before actually tearing it apart.

Binary star systems and the reduced mass

Two stars orbiting each other is a binary system. Unlike the Earth–Sun case (where the Sun is so massive that it barely wiggles), a typical binary has two comparable masses, and both orbit the shared centre of mass. Half of all stars in the Milky Way are actually members of binary or multi-star systems.

The centre of mass as the pivot

Let the two stars have masses m_1 and m_2, with position vectors \vec{r}_1 and \vec{r}_2 measured from some fixed external point. The centre of mass sits at

Why: this is just the definition of the centre of mass — a weighted average of the positions, weighted by mass.

With no external forces, \vec{R}_{\text{cm}} moves at constant velocity (or, if the system as a whole is at rest, it stays put). Work in the centre-of-mass frame: choose a frame moving with \vec{R}_{\text{cm}}, so the CM is permanently at the origin. Then

Why: with \vec{R}_{\text{cm}} = 0, the numerator of the CM definition vanishes, giving this linear relationship between the two position vectors.

Both stars orbit the CM, and they are always on opposite sides of it — so they orbit in sync, each tracing its own ellipse, with the CM at a common focus. The heavier star's ellipse is smaller (smaller \vec{r}_1 if m_1 > m_2), the lighter star's is larger.

Reducing the two-body problem to one body

Here is the slick move. Define the relative position

which is the vector from star 1 to star 2. This is the only separation that gravity cares about — the force on each star depends only on |\vec{r}|.

The force on star 1 due to star 2 is

Why: gravity pulls star 1 toward star 2. The direction is +\hat{r} (pointing from star 1 toward star 2, the direction of \vec{r}). The magnitude is Gm_1 m_2/r^2 — Newton's law.

and the force on star 2 is equal and opposite:

Divide each equation by its mass to get accelerations, then subtract the first from the second:

Why: subtracting the two equations of motion eliminates the centre-of-mass drift and leaves only the relative motion. Pull the common factor Gm_1 m_2/r^2 out and regroup the reciprocal masses.

The left side is \ddot{\vec{r}} — the second derivative of the relative position. Combine the two reciprocals on the right:

So

Why: the factor of m_1 m_2 cancels between numerator and denominator, leaving only m_1 + m_2.

Multiply both sides by the reduced mass

to get

Why: \mu \cdot G(m_1+m_2)/r^2 = m_1 m_2/(m_1+m_2) \cdot G(m_1+m_2)/r^2 = Gm_1m_2/r^2. The (m_1+m_2) in the numerator of G(m_1+m_2) cancels the (m_1+m_2) in the denominator of \mu, leaving a clean m_1 m_2.

This is the punchline. The equation \mu\, \ddot{\vec{r}} = -Gm_1 m_2 / r^2 \cdot \hat{r} looks exactly like Newton's second law for a single body of mass \mu, moving under the force of a single stationary mass (m_1 + m_2). The two-body problem is mathematically identical to a one-body problem with mass \mu orbiting a fixed gravitational source of mass (m_1+m_2).

The reduced mass itself

The reduced mass has an intuitive character:

• Equal masses (m_1 = m_2 = m): \mu = m^2/(2m) = m/2. Each body contributes half.
• Hugely unequal (m_1 \gg m_2): \mu \approx m_1 m_2 / m_1 = m_2. The reduced mass is just the lighter one. This is why the Earth–Sun problem reduces to "Earth orbits a fixed Sun": with m_{\odot} \gg m_{\oplus}, the Sun barely moves, and \mu \approx m_{\oplus}.

Once you have \mu, every orbital calculation — Kepler's third law, orbital energy, angular momentum — can be written for the reduced one-body problem. For instance, Kepler's third law for a binary becomes

where a is now the semi-major axis of the relative orbit. Measuring T and a gives you the total mass (m_1 + m_2) — one of the most important measurements in stellar astrophysics, and the foundation on which the mass–luminosity relation rests.

Tidal forces — the qualitative picture of ocean tides

Why does the Arabian Sea rise and fall at Mumbai twice a day? The naive answer is "the Moon pulls the water toward it." That gives one bulge, facing the Moon. It cannot explain the second tide twelve hours later.

The real answer is that tides are produced not by the Moon's gravity, but by differences in the Moon's gravity from one side of Earth to the other.

The tidal argument

Place the Earth (centre at O) a distance d from the Moon (mass M_m). Consider three points along the Earth–Moon line: the near side N, the centre C, and the far side F. Let Earth's radius be R_E \approx 6.37 \times 10^6 m and the Moon's distance d \approx 3.84 \times 10^8 m, so R_E / d \approx 0.017 — small but not zero.

The Moon's gravitational pull per kilogram (field strength) at each point:

Why: Newton's law of gravitation — field strength depends on distance squared. The near side is closer, the far side is farther.

All three point toward the Moon (call that direction positive). Now — and this is the key step — every point on Earth is in free fall toward the Moon at acceleration g_C (that is what keeps the Earth–Moon system orbiting their common centre of mass). So in the reference frame of the Earth's centre, the "extra" field experienced at any point is g - g_C:

At near side: g_N - g_C = GM_m\left[\dfrac{1}{(d-R_E)^2} - \dfrac{1}{d^2}\right] > 0, pointing toward the Moon.

At far side: g_F - g_C = GM_m\left[\dfrac{1}{(d+R_E)^2} - \dfrac{1}{d^2}\right] < 0, pointing away from the Moon.

Why: at the near side, the Moon pulls harder than the centre — so water there is pulled forward relative to Earth's centre. At the far side, the Moon pulls less than the centre — so the centre is pulled away from the water there, equivalent to the water being pushed outward (away from the Moon) relative to the centre. Both ends stretch outward from the centre.

This is the source of the two bulges. The ocean stretches outward on both the near and far sides. Earth rotates under this stretched water envelope, so any coastal location passes through two high tides per day — one when the Moon is overhead, one when the Moon is on the exact opposite side of the Earth.

Why the tidal effect scales as 1/d^3, not 1/d^2

This is a beautiful consequence. Expand 1/(d - R_E)^2 for R_E \ll d:

Why: use the binomial approximation (1+x)^{-2} \approx 1 - 2x for small x. Here x = -R_E/d, so (1 - R_E/d)^{-2} \approx 1 + 2R_E/d.

So the tidal (differential) field at the near side is

The tidal force falls off as 1/d^3, not 1/d^2. That extra factor of 1/d is why the Sun — which has a gravitational pull on Earth's oceans many times stronger than the Moon's — produces tides roughly half the size of the Moon's. The Sun is much farther, and 1/d^3 punishes distance severely.

When the Sun, Earth, and Moon line up (full or new moon), the two tidal effects add and you get spring tides — unusually high. When they are perpendicular (first or third quarter), they partially cancel and you get neap tides — unusually low. Fishermen along the Konkan and Andhra coasts use a tide table built directly from this physics to plan when to go out.

Weightlessness in orbit is free fall, not absence of gravity

An astronaut aboard the International Space Station drifts inside the cabin. The pen she releases hovers in the air. A water droplet from a drinking tube floats as a perfect sphere. Watching this, it is natural to conclude that gravity has switched off 400 km above the Earth.

It has not. At 400 km altitude, the gravitational field is

Why: substitute Earth's mass, the altitude of the ISS (h = 4 \times 10^5 m), and Earth's radius into Newton's law. The result is only a little less than the 9.8 m/s² at the surface — gravity at ISS altitude is about 89% of its surface value.

That is 89% of the surface value. Gravity is alive and well. So what is going on?

The elevator thought experiment

Imagine you are standing on a scale in a lift. The scale reads your weight: W = mg. Now someone cuts the cable. Lift, scale, and you all start falling together at g. As the lift accelerates downward, the floor drops out from under you at exactly the rate gravity is pulling you down. You are no longer pressing on the scale. The scale reads zero. You float.

During that terrifying fall, gravity is still acting on you. You are not weightless because gravity vanished — you are weightless because everything around you is falling at the same rate you are, so there is no contact force to register your weight.

This is exactly what is happening aboard the ISS. The station is in free fall around the Earth. So is every astronaut inside. So is every pen, water droplet, and scrap of food. Nothing presses against anything, so no one feels heavy.

The orbit is a permanent fall that misses

Here is the mental model that ties it together. Throw a cricket ball horizontally from the top of the Eiffel Tower (or Qutub Minar — 73 m tall). It arcs down and hits the ground a few dozen metres away. Throw it faster — it lands farther. Throw it so fast that the Earth's surface curves away from it at the same rate it falls — and it never hits the ground at all. It falls forever, continuously missing the Earth. That is an orbit. That is why the ISS stays up: it is falling at 8.7 \text{ m/s}^2, but moving sideways at 7.67 km/s, so the Earth's surface curves away from under it just as fast as it falls.

Gravity has not turned off. It is the thing that makes the orbit an orbit. Weightlessness is simply the signature that you, the station, and everything inside the station are all obeying that same gravitational acceleration — so nothing pulls on anything relative to anything else.

This picture, due to Newton's own thought experiment in the Principia, is the single cleanest answer to "what is orbit?" It is gravity, with enough sideways speed to make the falling infinite.

The Chandrayaan trajectory — gravity assists qualitatively

On 14 July 2023, ISRO's Chandrayaan-3 launched from Sriharikota. Six weeks later, its lander Vikram touched down gently near the Moon's south pole. The mission used a carefully shaped trajectory to get there on a modest fuel budget — raising its orbit around Earth with a series of small engine burns, letting Earth's gravity do most of the work, and finally letting the Moon capture it.

The trick at the heart of such trajectories is the gravity assist (also called a slingshot or flyby). It is how Voyager escaped the solar system, how Cassini reached Saturn, and how Chandrayaan extracted extra orbital energy without burning more fuel. The principle is surprising but clean.

The elastic-collision reframe

A spacecraft flying past a planet exchanges energy with it in a way that is, from the planet's rest frame, elastic. The spacecraft arrives at some speed v relative to the planet, loops around, and departs at the same speed v relative to the planet — the direction has changed, but not the magnitude. From the planet's frame, nothing is gained or lost.

But the planet is not standing still. It is moving around the Sun at some speed V_p. Switch to the Sun's frame and add \vec{V}_p to every velocity. Now the spacecraft arrived at \vec{v}_{\text{in}} + \vec{V}_p and left at \vec{v}_{\text{out}} + \vec{V}_p. Because the spacecraft swung its velocity vector around during the flyby, the magnitude of its velocity in the Sun's frame has changed — it can be faster (gravity assist forward) or slower (gravity brake).

This is why gravity assists work: the spacecraft plays elastic-collision games with a moving planet, and the planet's motion gets transferred. A deep flyby of Jupiter can boost a small probe by several kilometres per second — a speed change that would take tons of fuel to accomplish with rockets.

Chandrayaan's shape

Chandrayaan-3 did not need the big gravity assists that Voyager used. It used a lower-budget trick: repeated perigee burns. Starting in a low elliptical orbit around Earth, the engines fired at perigee (the point closest to Earth, where kinetic energy is highest). Each small burn raised the apogee — the far end of the ellipse — without changing perigee much, because firing at perigee converts the fuel into orbital energy most efficiently. Over five or six such burns, the apogee grew until it intersected the Moon's orbit, and Chandrayaan coasted outward on an ellipse that brought it into the Moon's gravitational reach. The Moon then captured it.

The reason perigee burns are so efficient is a subtle consequence of orbital energy. At perigee, the spacecraft is moving fastest, so the same \Delta v burn produces a larger \Delta(\tfrac{1}{2}v^2) — more work per kilogram of fuel. This is called the Oberth effect, and it is why ISRO's mission planners always schedule major burns at perigee. The physics is just W = \vec{F}\cdot\vec{v} — force times velocity — writ large across the solar system.

Worked examples

Common confusions

• "Self-energy is the energy inside the sphere." Self-energy is not a density stored somewhere. It is the total work done by gravity during the sphere's assembly (or equivalently, the negative of the work needed to disperse it). It is a property of the configuration, not a substance inside it.

• "The 3/5 coefficient is universal." No. 3/5 is specific to a uniform sphere. Centrally concentrated distributions (real stars) have coefficients between 3/5 and 1, depending on their density profile. For a polytropic star model with index n, the coefficient is \frac{3}{5-n}. Use 3/5 only when uniform density is specified or is a good approximation.

• "Reduced mass is the average of the two masses." No. \mu = m_1 m_2/(m_1+m_2) is the harmonic-mean-like combination. For equal masses m, it is m/2 (half the common mass, not the full mass or the sum). Always check with a limit: m_1 \to \infty should give \mu \to m_2, which \mu = m_1 m_2/(m_1+m_2) does.

• "Tides are caused by the Moon pulling the water up." Only half-true. The Moon does pull the near-side ocean up, but the far-side ocean also bulges outward, in the opposite direction. Tides are a differential (tidal) effect — the spatial variation of the Moon's pull across Earth's diameter — not just a net pull.

• "Astronauts float because there's no gravity in space." Wrong. Gravity at ISS altitude is 89% of its surface value. Astronauts float because the station and its contents are all in free fall together — the entire freely falling system has no contact forces. Gravity is what keeps the station in orbit; it has not disappeared.

• "A gravity assist gives the spacecraft energy for free." It steals energy from the planet. The planet slows in its solar orbit by a tiny, unmeasurable amount; the spacecraft speeds up by kilometres per second. Total energy of (planet + spacecraft) is conserved, but the distribution shifts dramatically because the planet is so much heavier.

Where this leads next

• Gravitational Potential and Potential Energy — where U = -GMm/r comes from and why the sign matters. The foundation for everything in this article.
• Satellites and Kepler's Laws — the three laws, applied to circular and elliptical orbits. The binary-star Kepler's law used above is derived there for the one-body case.
• Escape Velocity and Orbital Velocity — v_{\text{esc}} = \sqrt{2GM/R} and v_{\text{orb}} = \sqrt{GM/r}. Key inputs to gravity assists and transfer orbits.
• Elastic Collisions — the physics underlying gravity assists (in the planet's reference frame, a flyby is an elastic collision).
• Centre of Mass — the pivot point for the binary reduction, and the key to understanding why both stars orbit a common focus.