Escape Velocity

In 1783, a Yorkshire clergyman named John Michell wrote a letter that ended with a monster. Take the escape-speed idea seriously, he argued — every star has a speed a projectile must reach to leave it forever — and imagine a star so massive that this speed exceeds the speed of light itself. "All light emitted from such a body would be made to return towards it." An invisible star: the first black hole, a century and a half before general relativity, reached with nothing but the physics in this article. The everyday version of Michell's number is the most famous velocity in spaceflight: $11.2\ \mathrm{km/s}$ , Earth's escape speed. The previous article dug the gravity well and priced its full depth at about $62.5$ megajoules per kilogram; this article cashes that cheque. Escape velocity is nothing more than the launch speed whose kinetic energy exactly pays the well's full depth — one energy equation, two snapshots, no trajectory ever solved. That economy is why examiners love it: CBSE and ISC ask for the energy derivation, NEET asks what escape speed does not depend on, and JEE dresses the same equation in planets of doubled radius, launches at half the escape speed, and speeds left over at infinity. By the end you will own the formula $v_{\rm esc} = \sqrt{2GM/R} = \sqrt{2gR}$ , the $\sqrt2$ that separates orbiting from leaving, and the two-snapshot method that solves every "does it get away?" question ever set.

Key takeaways

Escape velocity is the minimum launch speed from a body's surface that never falls back: kinetic energy at launch exactly equals the well depth, $\tfrac12 mv_{\rm esc}^2 = \dfrac{GMm}{R}$ .
The result: $v_{\rm esc} = \sqrt{\dfrac{2GM}{R}} = \sqrt{2gR}$ — Earth $11.2\ \mathrm{km/s}$ , Moon $2.37\ \mathrm{km/s}$ .
It does not depend on the projectile's mass or on the direction of launch — energy is a scalar and $m$ cancels.
The $\sqrt2$ ladder: at any radius, escape speed is exactly $\sqrt2$ times circular-orbit speed, because escape pays the full well depth while an orbit carries only half.
Launched slower than $v_{\rm esc}$ , the projectile climbs to a finite $r_{\max}$ ; faster, it keeps $v_\infty = \sqrt{v_0^2 - v_{\rm esc}^2}$ forever. The sign of $E = \tfrac12 mv^2 - \tfrac{GMm}{r}$ decides everything.
Key idea: escape is an energy transaction — reach total energy zero and the well can never pull you back.

The core insight

The escape velocity from radius $R$ around mass $M$ is $v_{\rm esc} = \sqrt{2GM/R}$ : the speed at which kinetic energy exactly cancels gravitational potential energy, making the total energy zero — the boundary between bound and free, independent of the projectile's mass and of the direction of launch.

Where this lives today

No rocket ever launches at $11.2\ \mathrm{km/s}$ — thrust-as-you-go means real ascent profiles stay far slower in the lower atmosphere — but mission designers live on this article's ledger. Interplanetary probes are specified by their $C_3 = v_\infty^2$ , literally the "energy left over after the well," the quantity in Derivation B's Step 4; a Mars window needs $C_3 \approx 12\ \mathrm{km^2/s^2}$ , and you can now read that as $v_0$ just $0.5\ \mathrm{km/s}$ above escape speed at burnout. The $\sqrt2$ ladder is a budgeting tool: from low Earth orbit at $7.8\ \mathrm{km/s}$ , escaping costs only $\sim3.2\ \mathrm{km/s}$ more — which is why missions refuel their ambitions in orbit first. At the other extreme, Michell's limit is now observational science: the Event Horizon Telescope's images of M87 and Sagittarius A photograph the boundary where $v_{\rm esc}$ reaches $c$ . And closer to home, the leaky-planet argument of worked example 2 is live climate history — it is how Mars, with $v_{\rm esc} = 5.0\ \mathrm{km/s}$ , lost most of its air, a loss NASA's MAVEN mission still watches in real time.

The physical picture

Picture the gravity well from the previous article: a funnel-shaped energy landscape, $U = -GMm/r$ , steep near the surface and flattening forever outward. Standing on Earth's surface you sit at depth $GMm/R$ below the rim — about $62.5$ megajoules for every kilogram of you. The well's crucial feature, proved there, is that this depth is finite even though gravity's reach is infinite: the $1/r^2$ force weakens fast enough that the infinite climb costs a finite fare. Finite fare means a finite speed can pay it. Give a projectile kinetic energy equal to the full well depth and it can coast up the entire infinite slope, arriving "at the rim" — infinitely far away — with exactly nothing left. Give it one joule less and somewhere out there it runs dry, stops, and falls back: gravity never switches off, so any leftover debt is eventually collected. Give it one joule more and it not only leaves but keeps the surplus as speed forever. Escape velocity is the break-even point of this transaction, and that is why the whole topic is one line of energy bookkeeping.

Figure

The gravity well and the escape transaction — pay the full depth, or fall back

Two readings of the picture before the algebra. First, direction never enters. The well is an energy landscape, and energy has no arrows: launch straight up, at $45°$ , or sideways — as long as the path misses the planet, the same speed buys the same escape. "Escape velocity" is really an escape speed. Second, the picture shows why orbiting and escaping are cousins: a circular orbit at the surface radius carries kinetic energy $GMm/2R$ — exactly half the well depth (gravity supplies centripetal force, and that condition fixes $v^2 = GM/R$ ). Escape demands the full depth. The ratio of the speeds is therefore $\sqrt2$ , a factor that follows every astronaut from launch to the outer planets. One honest caution: this whole story prices the journey against gravity alone. Real rockets also fight air drag and, more importantly, do not need to start at escape speed at all — they thrust continuously, which is why no astronaut has ever travelled at $11.2\ \mathrm{km/s}$ relative to the ground. Escape velocity answers a cleaner question: with what speed must an unpowered projectile leave the surface to never return?

The math: one equation, two snapshots

Everything runs on the total mechanical energy of a projectile of mass $m$ at distance $r$ from the centre of a body of mass $M$ , moving at speed $v$ . Gravity is the only force, and it is conservative, so this number is locked at launch:

E = \frac12 mv^2 - \frac{GMm}{r} \qquad (1)

where: $\tfrac12mv^2$ — kinetic energy, always positive; $-GMm/r$ — the well from the previous article, always negative; $E$ — the constant that decides the projectile's fate. If $E < 0$ the kinetic energy dies before the potential energy does: there is a largest reachable $r$ , and the projectile is bound. If $E \ge 0$ it can afford the whole climb: free. Every question in this topic is equation (1) written at two moments and set equal. The escape condition is the marginal case — arrive at infinity with nothing to spare — and solving it (Derivation A) gives the master result:

v_{\rm esc} = \sqrt{\frac{2GM}{R}} = \sqrt{2gR} \qquad (2)

where: $R$ — the launch radius (the surface, for a ground launch); $g = GM/R^2$ — the surface gravity, which supplies the second form. Read the structure: the mass $m$ of the projectile has cancelled — a pebble and a battleship escape at the same speed — and no angle appears anywhere. For Earth, $\sqrt{2 \times 9.8 \times 6.37\times10^6} \approx 1.12\times10^4\ \mathrm{m/s} = 11.2\ \mathrm{km/s}$ ; for the Moon, $2.37\ \mathrm{km/s}$ — which is why leaving the Moon took a single small engine while leaving Earth took a Saturn V. For ratio questions — the exam's favourite costume — substitute $M = \tfrac43\pi R^3\rho$ into (2):

v_{\rm esc} = R\,\sqrt{\frac{8\pi G\rho}{3}} \qquad (3)

where: $\rho$ — the planet's mean density. At fixed density, escape speed grows linearly with radius; at fixed radius it grows as $\sqrt\rho$ . This one substitution answers the entire "planet X has twice the radius and the same density" family in a line.

you are asked	reach for	note
escape speed from a surface	$v_{\rm esc} = \sqrt{2GM/R} = \sqrt{2gR}$ , eq. (2)	mass-free, direction-free
compare two planets	ratios of (2), or eq. (3) with $\rho$	most MCQs need no numbers
launched slower — how high?	eq. (1) at launch and at the top ( $v=0$ )	never $v^2/2g$ unless $h \ll R$
launched faster — speed far away?	$v_\infty = \sqrt{v_0^2 - v_{\rm esc}^2}$	Pythagoras of escape
escape from an orbit at $r$	$v_{\rm esc}(r) = \sqrt2\,v_{\rm circ}(r)$	the $\sqrt2$ ladder

Derivation AEscape velocity by two energy snapshots

We derive equation (2): the minimum speed, launched from radius $R$ around mass $M$ , that never falls back. The strategy is the energy method at its purest — write the total energy (1) at launch, write it again in the marginal escaped state, and equate; no trajectory is solved because none is needed. The point to watch is what "minimum" means: barely escaping is arriving infinitely far away with zero speed left, and that single word "barely" is what pins the total energy to exactly zero.

Step 1. Snapshot one — launch. Speed $v_{\rm esc}$ at distance $R$ from the centre:

E = \frac12 mv_{\rm esc}^2 - \frac{GMm}{R}

Step 2. Snapshot two — the marginal escape. At $r \to \infty$ the potential energy $-GMm/r \to 0$ , and "barely made it" means the speed has also dwindled to zero, so the kinetic energy $\to 0$ too:

E = 0 + 0 = 0

Why this is the minimum: arrive at infinity with speed to spare and the launch was faster than necessary; stop at any finite $r$ and it was too slow — gravity still pulls there, and the projectile falls back. Zero-at-infinity is exactly the break-even.

Step 3. Energy is conserved between the snapshots, so the two expressions are the same number:

\frac12 mv_{\rm esc}^2 = \frac{GMm}{R} \quad\Longrightarrow\quad v_{\rm esc} = \sqrt{\frac{2GM}{R}}

Why $m$ vanished: it multiplies both sides — every kilogram carries its own fare and its own ticket. This cancellation is the same one that makes all objects fall alike.

Step 4. Convert to the surface-gravity form. Since $g = GM/R^2$ , we have $GM = gR^2$ , so

v_{\rm esc} = \sqrt{\frac{2gR^2}{R}} = \sqrt{2gR}

Why examiners love this form: no $G$ , no planetary mass — just the two numbers everyone knows. For Earth: $\sqrt{2 \times 9.8 \times 6.4\times10^6} = \sqrt{1.2544\times10^8} = 11\,200\ \mathrm{m/s}$ . *Verified: limits — a deeper well (larger $M$ , smaller $R$ ) raises $v_{\rm esc}$ ✓; $M \to 0$ gives $v_{\rm esc} \to 0$ (nothing to escape) ✓; Moon numbers in worked example 2 give $2.37\ \mathrm{km/s}$ , matching NCERT's quoted value ✓.*

Go deeper · Derivation B · 1 minThe √2 ladder — orbiting, escaping, and everything between

Derivation BThe √2 ladder — orbiting, escaping, and everything between

We derive the relation $v_{\rm esc} = \sqrt2\,v_{\rm circ}$ at any radius, and the two payoff formulas for launches that miss the mark: the maximum height for $v_0 < v_{\rm esc}$ and the leftover speed for $v_0 > v_{\rm esc}$ . The strategy is to keep using equation (1) at two snapshots and let the sign of $E$ tell the story. Watch how three apparently separate exam questions fall out of the same two lines of algebra.

Step 1. Circular-orbit speed at radius $r$ : gravity supplies the centripetal force, $\dfrac{GMm}{r^2} = \dfrac{mv_{\rm circ}^2}{r}$ , so

v_{\rm circ} = \sqrt{\frac{GM}{r}}

Step 2. Compare with escape from the same radius, $v_{\rm esc}(r) = \sqrt{2GM/r}$ :

\frac{v_{\rm esc}}{v_{\rm circ}} = \sqrt{2}

Why: an orbit's kinetic energy is half the local well depth (Step 1 says $\tfrac12mv_{\rm circ}^2 = GMm/2r$ ); escape needs the full depth. Half versus whole is the entire content of the $\sqrt2$ .

Step 3. Launched too slow — $v_0 < v_{\rm esc}$ , radially upward. Energy at launch equals energy at the top ( $v = 0$ , distance $r_{\max}$ ):

\frac12 v_0^2 - \frac{GM}{R} = -\frac{GM}{r_{\max}} \quad\Longrightarrow\quad r_{\max} = \frac{R}{1 - \dfrac{v_0^2}{v_{\rm esc}^2}}

Why the compact form: divide through by $GM/R$ and recognise $v_{\rm esc}^2 = 2GM/R$ . A quick check — $v_0 = v_{\rm esc}/2$ gives $r_{\max} = R/(1 - \tfrac14) = \tfrac43 R$ , a height of $R/3$ above the ground.

Step 4. Launched too fast — $v_0 > v_{\rm esc}$ . Same equation, far-away snapshot now has speed $v_\infty$ :

\frac12 v_0^2 - \frac{GM}{R} = \frac12 v_\infty^2 \quad\Longrightarrow\quad v_\infty = \sqrt{v_0^2 - v_{\rm esc}^2}

Why it looks like Pythagoras: the well claims a fixed energy toll $\tfrac12 v_{\rm esc}^2$ per kilogram; whatever kinetic energy remains is yours to keep, forever. *Verified: $v_0 = v_{\rm esc}$ in Step 3 gives $r_{\max} \to \infty$ ✓ and in Step 4 gives $v_\infty = 0$ ✓ — the two formulas meet exactly at the escape condition. $v_0 = 2v_{\rm esc}$ leaves $v_\infty = \sqrt3\,v_{\rm esc}$ , used in worked example 4.*

Figure

The speed ladder at Earth's surface — where a launch ends up

Limiting cases

case	result	meaning
$v_0 = v_{\rm esc}$	$E = 0$ , arrives at infinity with $v = 0$	the marginal, parabolic escape
$v_0 < v_{\rm esc}$	stops at $r_{\max} = R/(1 - v_0^2/v_{\rm esc}^2)$ , falls back	bound: $E < 0$
$v_0 > v_{\rm esc}$	keeps $v_\infty = \sqrt{v_0^2 - v_{\rm esc}^2}$ forever	free with surplus: $E > 0$
$v_0 \ll v_{\rm esc}$ , small heights	$r_{\max} - R \approx v_0^2/2g$	the familiar $mgh$ world, recovered
same $\rho$ , radius doubled	$v_{\rm esc}$ doubles	eq. (3): $v_{\rm esc} \propto R\sqrt\rho$
launch from orbit radius $r$	$v_{\rm esc}(r) = \sqrt2\,v_{\rm circ}(r)$	the $\sqrt2$ ladder, any rung

Worked examples

1 · Board level — the formula, cleanly (CBSE/ISC pattern, 3 marks)

Calculate the escape velocity from the Earth's surface, given $g = 9.8\ \mathrm{m/s^2}$ and $R = 6.4\times10^6\ \mathrm{m}$ . State two quantities on which your answer does not depend.

Step 1. Choose the surface-gravity form of equation (2) — it needs no $G$ and no planetary mass: $v_{\rm esc} = \sqrt{2gR}$ .

Step 2. Substitute: $v_{\rm esc} = \sqrt{2 \times 9.8 \times 6.4\times10^6} = \sqrt{1.2544\times10^8}$ .

Step 3. Evaluate: $v_{\rm esc} = \boxed{1.12\times10^4\ \mathrm{m/s} = 11.2\ \mathrm{km/s}}$ .

Step 4. Independence (the stated marks): the escape speed does not depend on the mass of the projectile (it cancels in the energy equation) or on the direction of launch (energy is a scalar). Writing "because energy conservation contains speed, not velocity" earns the reasoning mark.

2 · Board level — another world (ISC/NEET pattern)

The Moon has mass $7.35\times10^{22}\ \mathrm{kg}$ and radius $1.74\times10^6\ \mathrm{m}$ . Find its escape speed, and explain in one sentence why the Moon has essentially no atmosphere. ( $G = 6.67\times10^{-11}\ \mathrm{N\,m^2/kg^2}$ .)

Step 1. Compute $GM$ first — calculator hygiene: $GM = 6.67\times10^{-11} \times 7.35\times10^{22} = 4.90\times10^{12}\ \mathrm{m^3/s^2}$ .

Step 2. Apply equation (2): $v_{\rm esc} = \sqrt{\dfrac{2 \times 4.90\times10^{12}}{1.74\times10^6}} = \sqrt{5.63\times10^6}$ .

Step 3. Evaluate: $v_{\rm esc} = \boxed{2.37\times10^3\ \mathrm{m/s} \approx 2.4\ \mathrm{km/s}}$ — about a fifth of Earth's.

Step 4. The atmosphere sentence: gas molecules at lunar daytime temperatures have thermal speeds within striking distance of $2.4\ \mathrm{km/s}$ (fast molecules in the distribution exceed it), so over geological time any atmosphere leaks to space — a low escape speed is a leaky planet.

3 · Competitive level — scaling, no numbers (JEE-Main pattern, ~2 minutes)

A planet has twice the Earth's radius and the same mean density. Its escape velocity is (a) $11.2$ , (b) $22.4$ , (c) $5.6$ , (d) $44.8\ \mathrm{km/s}$ .

Step 1. Reach for the density form, equation (3): $v_{\rm esc} = R\sqrt{8\pi G\rho/3}$ , so at fixed density $v_{\rm esc} \propto R$ .

Step 2. Radius doubles, density fixed: $v_{\rm esc} = 2 \times 11.2 = \boxed{22.4\ \mathrm{km/s}}$ — option (b).

Step 3. Sanity by the long route: $M \propto \rho R^3$ grows $8\times$ ; $v_{\rm esc} = \sqrt{2GM/R} \propto \sqrt{8/2} \cdot \sqrt{2G/1}$ … cleaner as $\sqrt{M/R} = \sqrt{8/2} = 2$ ✓. Timing note: the moment a problem says "same density," convert to equation (3) mentally — $v_{\rm esc} \propto R\sqrt\rho$ — and answer in under thirty seconds. Deriving from $\sqrt{2GM/R}$ each time costs a minute you don't have. If instead the problem fixes $M$ and halves $R$ , use $v_{\rm esc} \propto \sqrt{M/R}$ : factor $\sqrt2$ .

4 · Competitive level — under and over the bar (JEE-Advanced flavour, ~4 minutes)

A projectile is fired vertically from Earth's surface (radius $R$ ), drag neglected. (a) If its launch speed is half the escape velocity, show that it rises to a height $R/3$ . (b) If instead it is launched at twice the escape velocity, find its speed when it is very far from Earth, in km/s.

Step 1. Part (a), two snapshots of equation (1) — launch and the top ( $v = 0$ , distance $r$ from centre): $\tfrac12 v_0^2 - \dfrac{GM}{R} = -\dfrac{GM}{r}$ .

Step 2. Insert $v_0 = \tfrac12 v_{\rm esc}$ , i.e. $v_0^2 = \tfrac14 \cdot \dfrac{2GM}{R} = \dfrac{GM}{2R}$ : the left side becomes $\dfrac{GM}{4R} - \dfrac{GM}{R} = -\dfrac{3GM}{4R}$ .

Step 3. So $\dfrac{GM}{r} = \dfrac{3GM}{4R}$ , giving $r = \dfrac{4R}{3}$ and height $h = r - R = \boxed{R/3 \approx 2100\ \mathrm{km}}$ . (The flat-Earth formula $v_0^2/2g = R/4$ underestimates — it keeps charging full gravity on a climb where the field weakens.)

Step 4. Part (b), the Pythagoras of escape (Derivation B, Step 4): $v_\infty = \sqrt{v_0^2 - v_{\rm esc}^2} = \sqrt{4v_{\rm esc}^2 - v_{\rm esc}^2} = \sqrt3\,v_{\rm esc}$ .

Step 5. Numbers: $v_\infty = 1.732 \times 11.2 = \boxed{19.4\ \mathrm{km/s}}$ . Timing note: both parts are the same two-snapshot template — budget ninety seconds each once you stop re-deriving. The traps are inserting $v_{\rm esc}/2$ as " $v_{\rm esc}^2/2$ " instead of $v_{\rm esc}^2/4$ , and answering part (b) with $2v_{\rm esc} - v_{\rm esc} = 11.2$ : speeds do not subtract linearly, their squares do.

Exam traps

$mgh$ smuggled into a planetary climb. The formula $h = v_0^2/2g$ is the $h \ll R$ zoom of the true energy equation. Example 4 shows the honest answer $R/3$ against the smuggled $R/4$ — a 25% error an examiner has planted options for. Rule: heights comparable to $R$ demand $-GMm/r$ bookkeeping.

Speeds subtracted linearly. "Launched at $2v_{\rm esc}$ , so it has $v_{\rm esc}$ left at infinity" — wrong; energies subtract, so squares of speeds subtract: $v_\infty = \sqrt{4 - 1}\,v_{\rm esc} = \sqrt3\,v_{\rm esc}$ .

The mass distractor. Every NEET season, options imply heavier rockets need higher escape speed. The speed is mass-free (the energy is not — that is why heavy rockets need more fuel, not more velocity).

The direction distractor. "Escape velocity applies only to vertical launch" — false; energy is direction-blind. What direction does change is whether the path clears the planet, and (for real launches) the free ride from Earth's rotation: launching eastward from the equator gifts $0.46\ \mathrm{km/s}$ of ground-relative speed, which is why spaceports hug the equator — a favourite one-liner in JEE options.

$\sqrt2$ confusions. Orbit speed at the surface radius is $7.9\ \mathrm{km/s}$ ; escape is $11.2 = \sqrt2 \times 7.9$ . MCQ options love offering $7.9$ , $11.2$ , $15.8$ ( $= \sqrt2 \times 11.2$ ) and $22.4$ together — know which rung of the ladder the question stands on: escaping from a circular orbit needs only $(\sqrt2 - 1)v_{\rm circ} \approx 0.41\,v_{\rm circ}$ extra.

Escape speed from altitude. From a height $h$ , use $R + h$ in equation (2), not $R$ — the well is shallower up there. The 11.2 km/s figure belongs to the surface alone.

Which $g$ , which $R$ . CBSE/ISC habitually supply $g = 9.8$ , $R = 6.4\times10^6\ \mathrm{m}$ (giving the clean $11.2$ ); JEE/NEET often say $g = 10$ , which gives $\sqrt{2\times10\times6.4\times10^6} = 11.3\ \mathrm{km/s}$ — do not "correct" it back to 11.2 if the paper's data disagree. Use what is given.

Origins · 1 minJohn Michell and how this was discovered

John Michell

1724–1793 · Thornhill, Yorkshire

John Michell was an eighteenth-century one-man research institution disguised as a village rector: he founded seismology (proposing earthquakes travel as waves), invented the torsion balance that his friend Henry Cavendish later used to weigh the Earth, and pioneered stellar statistics. His 1783 letter to Cavendish ran escape-velocity logic to its limit: since light was then thought to be particles slowed by gravity, a star 500 times the Sun's diameter at the same density would have an escape speed above light speed — "all light emitted from such a body would be made to return towards it." He even proposed detecting such dark stars by watching luminous companions wobble around an unseen partner — essentially how black holes and exoplanets are found today. The Newtonian argument mistreats light, yet the critical radius it implies, $R = 2GM/c^2$ , is exactly the Schwarzschild radius general relativity derived honestly in 1916. Michell's monster was the right answer from the wrong theory — physics' most productive coincidence.

History

1687

— Newton's Principia gives the force law; the finite depth of the

1/r

well makes escape a meaningful question.

1783

— Michell's letter to Cavendish: stars whose escape speed beats light would be invisible — the "dark star."

1796

— Laplace independently publishes the dark-star argument in Exposition du système du monde.

1865

— Jules Verne's From the Earth to the Moon fires a crewed shell at roughly escape speed — fatal for the crew, but the number is right.

1959

— Luna 1 becomes the first human artifact to exceed Earth's escape speed, sailing into a heliocentric orbit.

1969

— Apollo 11's ascent stage leaves the Moon on a single small engine — a 2.4 km/s well is a shallow one.

1977

— Voyagers 1 and 2 launch; gravity assists push them past the Sun's local escape speed, making them permanent interstellar objects.

Going further

Texts. NCERT Physics Class 11, Ch. 8 (Gravitation) · HC Verma, Concepts of Physics Vol. 1, Ch. 11 · Young & Freedman, Ch. 13.

Provenance. All results are standard canon. A verifier should recompute: $\sqrt{2 \times 9.8 \times 6.4\times10^6} = 11\,200\ \mathrm{m/s}$ ; the Moon's $GM = 4.90\times10^{12}$ and $v_{\rm esc} = \sqrt{5.63\times10^6} = 2.37\ \mathrm{km/s}$ ; example 3's factor 2 from $v \propto R\sqrt\rho$ ; example 4's $r_{\max} = 4R/3$ (from $v_0^2 = GM/2R$ ) and $v_\infty = \sqrt3 \times 11.2 = 19.4\ \mathrm{km/s}$ ; the ladder $11.2/7.9 = \sqrt2$ .

Prerequisites. Gravitational Potential Energy & Potential · Newton's Law of Universal Gravitation · Kinetic Energy & the Work–Energy Theorem. Leads to. Satellites: Speed, Period, Energy & Weightlessness · Raising & Switching Orbits (archetype) · Gravitational Potential & Escape Velocity (UG deep dive).

Why doesn't escape velocity depend on the direction of launch? Surely going straight up is easier than sideways?

The deciding equation is energy conservation, and energy contains speed squared — no direction anywhere. Whether you spend your kinetic energy climbing steeply or spiralling out gradually, the well charges the same total fare $GMm/R$ . The one real caveat: the path must not intersect the planet. Launch horizontally and "hitting the ground" is a geometry problem, not an energy one. (Real rockets pitch over and fly nearly horizontally quite quickly — for orbital mechanics reasons, not because the well cares.)

If gravity never ends, how can anything truly escape? Doesn't Earth eventually pull it back?

Gravity does keep pulling, forever — but pulling forever is not the same as winning. With $E \ge 0$ , the projectile's speed decreases along the whole journey yet never reaches zero: the deceleration weakens as $1/r^2$ so fast that the total speed loss is capped. Add up the infinite journey's toll and it is finite — the same finite-well-depth fact from the potential-energy article. Escape means gravity's remaining claim on you is smaller than the kinetic energy you still carry, all the way out.

Rockets never travel at 11.2 km/s, yet they leave Earth. Is the concept useless for real spaceflight?

Not useless — just precisely scoped. Escape velocity answers: "what speed must an unpowered projectile have at the surface?" A rocket under thrust rewrites the question continuously; in principle you could leave Earth at walking pace with an absurd fuel supply. But the energy verdict never changes — per kilogram delivered to deep space you must pay the same 62.5 MJ well fare, however you spread the payments. That is why engineers quote missions in Δv and $C_3 = v_\infty^2$ : this article's bookkeeping, in industry units.

Where does the √2 between orbital speed and escape speed come from?

From "half versus whole." A circular orbit at radius $r$ has gravity exactly supplying the centripetal force, which forces $\tfrac12mv^2 = GMm/2r$ — kinetic energy equal to half the local well depth. Escape requires kinetic energy equal to the full depth $GMm/r$ . Double the kinetic energy is $\sqrt2$ times the speed. This is also why a satellite needs only about 41% more speed to leave than to stay — low orbit really is "halfway to anywhere" in energy terms.

Why does a low escape speed mean a planet loses its atmosphere? The molecules aren't moving at kilometres per second… are they?

They are, some of them. At temperature $T$ , gas molecules move with a distribution of speeds; light molecules like hydrogen and helium have typical speeds of $1$ – $2\ \mathrm{km/s}$ at everyday temperatures, and the fast tail of the distribution reaches several times that. On Earth ( $11.2\ \mathrm{km/s}$ ) the tail that escapes is negligible except for hydrogen — which Earth does in fact leak. On the Moon ( $2.4\ \mathrm{km/s}$ ) essentially every gas has a fat escaping tail, so any atmosphere bleeds away in geologic time. Rule of thumb from planetary science: holding a gas for billions of years needs $v_{\rm esc}$ roughly six times that gas's typical thermal speed.

Is escape velocity the speed needed to "leave gravity" or to "leave the atmosphere"? My textbook seems to mix them up.

Neither is quite right, and the distinction is worth marks. Leaving the atmosphere (about 100 km up) costs a tiny fraction of the well fare — the Kármán line sits at $h/R \approx 0.016$ , barely a scratch on the 62.5 MJ/kg total. Escape velocity is about leaving the energy well entirely: reaching arbitrarily large distance. Between the two lives a third thing exams also probe — reaching orbit, which is mostly about gaining sideways speed ( $7.9\ \mathrm{km/s}$ at low altitude), not height. Three different feats, three different price tags.

Black holes: is Michell's dark-star argument actually correct, since it gets the Schwarzschild radius right?

The number is right; the physics is wrong twice, and the errors cancel. Michell treated light as particles that decelerate like stones — but light always moves at $c$ ; it loses energy by redshifting, not by slowing. And the Newtonian picture would let light rise, slow, and fall back — allowing distant observers brief glimpses — whereas a true event horizon is an absolute one-way surface. General relativity redoes the problem honestly and lands on the same $R = 2GM/c^2$ . Treat the Newtonian derivation as a mnemonic, not an explanation — saying so explicitly is a classic viva/interview point.