Kinetic Energy

In 1722 the Dutch physicist Willem 's Gravesande dropped brass balls into a bed of soft clay and measured the dents. Double the impact speed and the dent did not get twice as deep — it got four times deeper. Whatever "quantity of motion" the ball delivered to the clay, it grew with the square of the speed. Émilie du Châtelet seized on those dents to settle a fight that had been running since Leibniz: the energy of motion is not $mv$ , as Descartes' followers insisted, but the quantity we now write as $K = \tfrac12 mv^2$ — kinetic energy. That squared velocity is the most consequential exponent in everyday physics. It is why a car at 100 km/h needs four times the braking distance it needs at 50, why a rifle bullet weighing eight grams outdamages a thrown brick, and why every extra kilometre per second matters so desperately to a rocket engineer. And the formula hides a subtlety that surprises everyone once: kinetic energy is frame-dependent. The coffee in a train passenger's cup has zero kinetic energy to her and nearly a hundred joules to someone on the platform — and both are right. This article builds $\tfrac12 mv^2$ from the work integral (that is where the $\tfrac12$ comes from), splits the kinetic energy of any system into a center-of-mass part plus an internal part (König's theorem — the key to understanding collisions), and shows exactly how the bookkeeping survives a change of reference frame. Master this one scalar and the entire energy method of mechanics opens up.

Key takeaways

Kinetic energy is the scalar $K = \tfrac12 mv^2$ : the work needed to accelerate a mass $m$ from rest to speed $v$ , measured in joules.
The $\tfrac12$ and the $v^2$ are not conventions — they fall out of integrating Newton's second law along the path: $W_{\mathrm{net}} = \Delta K$ (the work–energy theorem).
$K$ forgets direction (a scalar) but remembers the frame: different inertial observers assign different kinetic energies to the same body, yet each observer's work–energy ledger balances perfectly.
Because $K \propto v^2$ , doubling speed quadruples the energy — and the braking distance, and the crash damage. Speed limits are quadratic-law policy.
For any system, König's theorem splits the total cleanly: $K = \tfrac12 M v_{\mathrm{cm}}^2 + K_{\mathrm{rel}}$ — bulk motion of the center of mass plus internal motion about it. Collisions can only ever spend the second part.
$\tfrac12 mv^2$ is the low-speed limit of the relativistic $K = mc^2(\gamma - 1)$ ; the correction reaches 1% only near 12% of the speed of light.
Key idea: kinetic energy $K = \tfrac12 mv^2$ is the scalar measure of motion whose change equals the net work done — it converts the vector language of forces into a single energy ledger.

The core insight

Kinetic energy $K = \tfrac12 mv^2$ is the scalar measure of motion whose change equals the net work done on a body — it converts the vector language of forces into a single energy ledger, and joins potential energy to form a conserved total when the forces allow it.

Where this lives today

Every electric vehicle on the road runs König-flavored energy bookkeeping in real time. A 1800 kg EV at 100 km/h carries about $\tfrac12(1800)(27.8)^2 \approx 695\ \mathrm{kJ} \approx 0.19\ \mathrm{kWh}$ of kinetic energy; regenerative braking runs the drive motor backward as a generator and pumps a sizable fraction of that back into the battery instead of boiling it off in brake discs. Over a stop-and-go commute the recovered $\tfrac12 mv^2$ is worth 15–30% of range — which is why city efficiency figures for EVs famously beat their highway figures, inverting a century of combustion-car intuition. At the other end of the scale, kinetic energy is now a planetary-defense tool. NASA's DART spacecraft — 570 kg at 6.1 km/s — deliberately slammed into the asteroid Dimorphos in 2022, delivering $\tfrac12 mv^2 \approx 1.1\times10^{10}\ \mathrm{J}$ , the energy of roughly 2.5 tons of TNT, from a body lighter than a small car. The impact shortened the moonlet's orbital period by 32 minutes, proving the "kinetic impactor" concept: at 6 km/s, velocity — entering squared — does the work that no practical amount of launched mass could. The same $v^2$ arithmetic governs the particle physicists' unit of choice: accelerator energies are quoted in electron-volts precisely because kinetic energy, not speed, is what a machine like the LHC (protons at 6.8 TeV, $\gamma \approx 7250$ ) actually purchases.

The physical picture

Push a stalled car down the street. The force you apply, accumulated over the distance the car rolls, does not vanish — it is banked in the car's motion. Kinetic energy is that bank account: net positive work fills it and the body speeds up; net negative work (brakes, friction, a hill) drains it; and when the account hits zero the body is momentarily at rest, at the turning point of its motion. The work–energy theorem is nothing but the bank statement: every joule deposited or withdrawn by the net force shows up, exactly, as a change in $\tfrac12 mv^2$ . Two features of this account deserve to be internalized before any algebra. First, it forgets direction: a 1 kg ball moving north at 3 m/s and an identical ball moving east at 3 m/s hold the same 4.5 J, because only the squared speed $v^2 = \mathbf{v}\cdot\mathbf{v}$ enters. That scalar blindness is a feature — it lets energy methods leap over the vector bookkeeping of forces whenever you only care about how fast, not which way. Second, it remembers the frame. Speed is not a property of a body alone; it is a relation between the body and an observer, and $K$ inherits that.

Figure

One cup of coffee, two honest ledgers — kinetic energy depends on who is measuring

The third habit to build is respect for the exponent. Because the account balance grows as $v^2$ , equal increments of speed are not equal increments of danger: going from 90 to 120 km/h adds far more energy than going from 60 to 90. A braking force limited by tire friction is roughly constant, so the distance needed to drain the account scales as $v^2$ too — double the speed, quadruple the stopping distance. Every road-safety table is a graph of a parabola.

Figure

The quadratic law of braking — stopping distance grows with the square of speed

The energy of motion, one formula at a time

Everything starts from the definition. For a single particle — any body whose internal structure we agree to ignore — the kinetic energy is built from the mass and the squared speed. The squared speed is best written as a dot product, because that form makes the scalar nature manifest and generalizes instantly to three dimensions:

K = \tfrac{1}{2}\,m\,v^2 = \tfrac{1}{2}\,m\,(\mathbf{v}\cdot\mathbf{v}) \qquad (1)

where: $K$ — kinetic energy (a scalar, in joules: $1\ \mathrm{J} = 1\ \mathrm{kg\,m^2/s^2}$ ); $m$ — the particle's mass; $\mathbf{v}$ — its velocity in the chosen inertial frame, with $v = |\mathbf{v}|$ . Notice what the formula does not contain: any reference to direction. A 0.16 kg cricket ball bowled at 40 m/s carries $\tfrac12(0.16)(40)^2 = 128\ \mathrm{J}$ whether it is moving north, east, or straight up — and it would carry a different number entirely to an observer driving alongside it. Often you know a body's momentum rather than its speed — in collisions, where momentum is the conserved currency, and throughout quantum mechanics, where momentum is the natural variable. Squaring $\mathbf{p} = m\mathbf{v}$ and dividing by $2m$ eliminates the velocity:

K = \frac{p^2}{2m} \qquad (2)

where: $p = |\mathbf{p}|$ — the magnitude of the momentum. This innocuous rewrite carries real physics: at equal momentum, the lighter body carries more kinetic energy. A 0.5 kg pistol and its 8 g bullet share equal-and-opposite momenta at firing, but the bullet, sixty times lighter, carries sixty times the kinetic energy — which is why the bullet does the damage and the shooter's hand merely stings. How fast does the account balance change? Differentiate (1) along the motion: $\mathrm{d}K/\mathrm{d}t = m\,\mathbf{v}\cdot\dot{\mathbf{v}} = \mathbf{F}_{\mathrm{net}}\cdot\mathbf{v}$ . The rate of change of kinetic energy is the power delivered by the net force:

P = \frac{\mathrm{d}K}{\mathrm{d}t} = \mathbf{F}_{\mathrm{net}}\cdot\mathbf{v} \qquad (3)

where: $P$ — power in watts; the dot product projects the force onto the velocity. Equation (3) is the sharpest diagnostic in mechanics: a force perpendicular to the velocity delivers zero power and cannot change the speed, only the direction. That single observation explains why the tension in a whirling string, the normal force on a sliding block, the magnetic force on a moving charge, and the Coriolis force in a rotating frame all leave kinetic energy untouched. Integrate (3) over time and you get the accumulated deposit — the work — and the central theorem of this article, whose full derivation is next:

W_{\mathrm{net}} = \int_{1}^{2}\mathbf{F}_{\mathrm{net}}\cdot\mathrm{d}\mathbf{r} = K_2 - K_1 = \Delta K \qquad (4)

where: $W_{\mathrm{net}}$ — work done by the net force along the actual trajectory from point 1 to point 2; $\mathrm{d}\mathbf{r}$ — an infinitesimal step along that trajectory. The theorem holds for any force law — conservative or not, constant or wildly varying — because it is nothing but Newton's second law integrated along the path.

Derivation AThe work integral builds one half m v squared

We are deriving the work–energy theorem (4) — and with it the very form $\tfrac12 mv^2$ — from nothing but Newton's second law. The strategy is to take $\mathbf{F} = m\,\mathrm{d}\mathbf{v}/\mathrm{d}t$ , dot it with a step $\mathrm{d}\mathbf{r}$ along the actual trajectory, and watch the right-hand side collapse into the derivative of a single scalar. This is where the mysterious factor of $\tfrac12$ enters: it is the same $\tfrac12$ that appears whenever you integrate something against its own rate of change, exactly as $\int x\,\mathrm{d}x = \tfrac12 x^2$ . The one step to watch is the product-rule identity in Step 2 — it is the hinge of the whole theorem.

Step 1. Dot Newton's second law with a displacement along the path, and trade the displacement for velocity times time, $\mathrm{d}\mathbf{r} = \mathbf{v}\,\mathrm{d}t$ :

\mathbf{F}\cdot\mathrm{d}\mathbf{r} = m\,\frac{\mathrm{d}\mathbf{v}}{\mathrm{d}t}\cdot\mathbf{v}\,\mathrm{d}t

Why this substitution is legal: we are integrating along the particle's actual trajectory, where position and time are locked together by the motion itself — $\mathrm{d}\mathbf{r}$ is not an arbitrary step but the step the particle really takes in $\mathrm{d}t$ .

Step 2. Recognize the integrand as a total derivative. By the product rule, $\tfrac{\mathrm{d}}{\mathrm{d}t}(\mathbf{v}\cdot\mathbf{v}) = 2\,\mathbf{v}\cdot\tfrac{\mathrm{d}\mathbf{v}}{\mathrm{d}t}$ , so:

\frac{\mathrm{d}\mathbf{v}}{\mathrm{d}t}\cdot\mathbf{v} = \tfrac12\,\frac{\mathrm{d}}{\mathrm{d}t}\big(\mathbf{v}\cdot\mathbf{v}\big) = \tfrac12\,\frac{\mathrm{d}(v^2)}{\mathrm{d}t}

Why the one half appears: differentiating $v^2$ produces a factor of 2 by the product rule, so undoing the derivative costs a factor of $\tfrac12$ . The half in $\tfrac12 mv^2$ is pure calculus, not physics.

Step 3. Integrate from point 1 to point 2. The integral of a total derivative depends only on the endpoint values:

W_{\mathrm{net}} = \int_{t_1}^{t_2} \tfrac12\,m\,\frac{\mathrm{d}(v^2)}{\mathrm{d}t}\,\mathrm{d}t = \tfrac12 m v_2^2 - \tfrac12 m v_1^2 = \Delta K

Why this defines kinetic energy: the combination $\tfrac12 mv^2$ is singled out by the dynamics itself — it is the unique function of speed whose endpoint difference equals the path integral of force. Nature chose the formula; we just named it.

Step 4. Check the constant-force special case. For a constant 1D force starting from rest, kinematics gives $v^2 = 2as$ , so $\tfrac12 mv^2 = m a s = Fs$ — force times distance, exactly as the theorem demands. As a consistency check: the general result reproduces $W = Fs$ for a constant force, and it is equivalent to the kinematic identity $v^2 = v_0^2 + 2as$ .

With the theorem in hand, the definition (1) stops being a convention and becomes a measurement: kinetic energy is the work a moving body can do on whatever stops it — the depth of 's Gravesande's dent. The family of formulas so far, and the situations that call for each:

you know…	reach for	why it wins there
the speed	$K = \tfrac12 mv^2$	the definition — direct
the momentum	$K = p^2/2m$	collisions and quantum mechanics conserve $p$ , not $v$
the force along a path	$W_{\mathrm{net}} = \Delta K$	converts dynamics into before/after bookkeeping
the force right now	$P = \mathbf{F}\cdot\mathbf{v}$	tells you if speed is rising or falling this instant
a rigid body's spin	$K_{\mathrm{rot}} = \tfrac12 I\omega^2$	every particle shares the same $\omega$
polar coordinates	$K = \tfrac12 m(\dot r^2 + r^2\dot\theta^2)$	orbits and central forces

The last two rows deserve their displays. A rigid body spinning about a fixed axis has every particle moving in a circle of its own radius $r_i$ at the shared angular speed $\omega$ , so summing $\tfrac12 m_i (r_i\omega)^2$ over the body factors into:

K_{\mathrm{rot}} = \tfrac12\,I\,\omega^2, \qquad I = \sum_i m_i r_i^2 \qquad (5)

where: $I$ — the moment of inertia about the rotation axis (kg·m²); $\omega$ — angular speed (rad/s). Same law, new inertia: $I$ plays the role of mass, $\omega$ the role of speed. A grid-scale flywheel stores megajoules precisely this way — mass pushed far from the axis, spun fast, because energy buys $\omega^2$ . And for a particle tracked in plane polar coordinates $(r, \theta)$ — the natural language of orbits — the velocity splits into a radial part $\dot r$ and a transverse part $r\dot\theta$ , which are perpendicular, so their squares add:

K = \tfrac12\,m\left(\dot r^2 + r^2\dot\theta^2\right) \qquad (6)

where: $\dot r$ — how fast the radius stretches; $r\dot\theta$ — how fast the particle sweeps sideways. In the Kepler problem this split becomes the whole story: the transverse term turns into the centrifugal barrier $L^2/2mr^2$ , and planetary motion reduces to one-dimensional energy bookkeeping.

Systems of particles: the König split

Real bodies are crowds: a thrown wrench is $10^{24}$ atoms translating, tumbling, and vibrating at once. Summing $\tfrac12 m_i v_i^2$ over the crowd looks hopeless until you ask the right question — how much of the total is bulk motion, going somewhere together, and how much is internal motion, going nowhere on average? The center of mass makes that split exact, and the cross term between the two contributions dies identically.

Figure

König's theorem — total kinetic energy splits into center-of-mass drift plus internal motion

The result is König's theorem, and it earns its display because it is the master key to collisions:

K = \tfrac12\,M\,v_{\mathrm{cm}}^2 \;+\; \underbrace{\tfrac12\sum_i m_i\,v_i'^2}_{K_{\mathrm{rel}}} \qquad (7)

where: $M = \sum_i m_i$ — total mass; $v_{\mathrm{cm}}$ — speed of the center of mass; $v_i'$ — speed of particle $i$ relative to the center of mass. The first term is the kinetic energy the system would have if all its mass rode at the CM; the second is everything else — tumbling, vibration, relative approach. The decisive physics: in an isolated collision, momentum conservation freezes $v_{\mathrm{cm}}$ , so the first term is untouchable. Only $K_{\mathrm{rel}}$ is available to be dissipated into heat, sound, and crumpled metal — and in a perfectly inelastic collision, all of it is.

Go deeper · Derivation B · 1 minKoenig's theorem — splitting the energy at the center of mass

Derivation BKoenig's theorem — splitting the energy at the center of mass

The goal is equation (7): the total kinetic energy of any system equals the kinetic energy of its center of mass plus the kinetic energy of motion about the center of mass, with no leftover piece. The strategy is transparent — write each particle's velocity as the CM velocity plus a residual, substitute into $\sum \tfrac12 m_i v_i^2$ , and expand the square. Everything hinges on the cross term, and the point of the derivation is to see that it vanishes not approximately but identically, by the very definition of the center of mass. This is a theorem about bookkeeping, so it holds for gases, rigid bodies, colliding carts, and galaxies alike.

Step 1. Decompose each velocity relative to the center of mass: $\mathbf{v}_i = \mathbf{v}_{\mathrm{cm}} + \mathbf{v}_i'$ , and substitute:

K = \tfrac12\sum_i m_i\,(\mathbf{v}_{\mathrm{cm}} + \mathbf{v}_i')\cdot(\mathbf{v}_{\mathrm{cm}} + \mathbf{v}_i')

Why this decomposition and not another: the CM is the unique point whose residual momenta sum to zero — any other reference velocity would leave a surviving cross term.

Step 2. Expand the dot product into three sums:

K = \tfrac12\,M v_{\mathrm{cm}}^2 \;+\; \mathbf{v}_{\mathrm{cm}}\cdot\sum_i m_i\mathbf{v}_i' \;+\; \tfrac12\sum_i m_i v_i'^2

Why the first term collapses: $\mathbf{v}_{\mathrm{cm}}$ is the same for every particle, so it factors out of the sum, leaving $\sum m_i = M$ .

Step 3. Kill the cross term. The total momentum measured relative to the CM is, by the definition of the CM position $\mathbf{R} = \sum m_i \mathbf{r}_i / M$ :

\sum_i m_i\mathbf{v}_i' = \sum_i m_i(\mathbf{v}_i - \mathbf{v}_{\mathrm{cm}}) = M\mathbf{v}_{\mathrm{cm}} - M\mathbf{v}_{\mathrm{cm}} = \mathbf{0}

Why this is exact and not an approximation: it is the derivative of the identity $\sum m_i \mathbf{r}_i' = M(\mathbf{R} - \mathbf{R}) = 0$ — in the CM frame, the mass-weighted average velocity is zero by construction.

Step 4. What survives is the theorem: $K = \tfrac12 M v_{\mathrm{cm}}^2 + K_{\mathrm{rel}}$ , with both pieces separately non-negative. The two-cart system of Worked Example 2 confirms the split numerically: total $40.5\ \mathrm{J} = 27.0\ \mathrm{J}$ (CM part, $\tfrac12\cdot 6\cdot 3^2$ ) $+\ 13.5\ \mathrm{J}$ (relative part) — the split is exact.

Changing frames without breaking the books

Here is the subtlety that the coffee-cup picture promised. Ride alongside a moving body and its kinetic energy drops to zero; watch from the platform and it is hundreds of joules. If observers disagree about $K$ itself, do they disagree about physics? They do not — and seeing exactly why is one of the most instructive calculations in mechanics. Boost to a frame moving at constant velocity $\mathbf{u}$ , so every velocity transforms as $\mathbf{v}' = \mathbf{v} - \mathbf{u}$ . Expanding the square gives the transformation law for a system's total kinetic energy:

K' = K - \mathbf{u}\cdot\mathbf{P} + \tfrac12\,M u^2 \qquad (8)

where: $\mathbf{P} = M\mathbf{v}_{\mathrm{cm}}$ — the system's total momentum in the original frame. Kinetic energy is not invariant, and it is not merely shifted by a constant either — the correction depends on the momentum, so even energy differences change between frames. The books still balance because the work transforms in exactly the same way, as the derivation shows. One elegant special case of (8): boost into the CM frame ( $\mathbf{u} = \mathbf{v}_{\mathrm{cm}}$ ) and the formula returns $K' = K - \tfrac12 Mv_{\mathrm{cm}}^2 = K_{\mathrm{rel}}$ — König's theorem rediscovered as the minimum kinetic energy any observer can assign to the system.

Go deeper · Derivation C · 1 minKinetic energy under a Galilean boost — the ledger survives

Derivation CKinetic energy under a Galilean boost — the ledger survives

We are deriving the transformation law (8) and then confronting the apparent paradox head-on: if $\Delta K$ differs between inertial frames, how can the work–energy theorem hold in both? The strategy is to transform the two sides of the theorem separately — the kinetic energy by direct substitution, the work by transforming the displacement in the work integral — and watch both sides shift by the same amount. That shared shift is the entire resolution: each observer keeps honest books; they just book different totals. Throughout, the boost velocity $\mathbf{u}$ is constant, so both frames are inertial and Newton's laws (with the same force $\mathbf{F}$ ) apply in each.

Step 1. Transform the kinetic energy of a particle. With $\mathbf{v}' = \mathbf{v} - \mathbf{u}$ :

K' = \tfrac12 m\,(\mathbf{v}-\mathbf{u})\cdot(\mathbf{v}-\mathbf{u}) = K - m\,\mathbf{u}\cdot\mathbf{v} + \tfrac12 m u^2

Why summing over particles gives (8): the middle term collects $\mathbf{u}\cdot\sum m_i\mathbf{v}_i = \mathbf{u}\cdot\mathbf{P}$ and the last collects $\tfrac12 u^2 \sum m_i = \tfrac12 M u^2$ .

Step 2. Take the change between two moments of a process. The $\tfrac12 mu^2$ term is constant and drops out, leaving:

\Delta K' = \Delta K - \mathbf{u}\cdot\Delta\mathbf{p} = \Delta K - \mathbf{u}\cdot\mathbf{J}

Why $\Delta\mathbf{p}$ appears: the momentum change equals the impulse $\mathbf{J} = \int\mathbf{F}\,\mathrm{d}t$ delivered by the net force — so the frame disagreement about energy change is controlled entirely by the impulse.

Step 3. Now transform the work. Forces and time intervals are Galilean invariants, but displacements are not: $\mathrm{d}\mathbf{r}' = \mathbf{v}'\,\mathrm{d}t = (\mathbf{v}-\mathbf{u})\,\mathrm{d}t$ . Therefore:

W' = \int \mathbf{F}\cdot\mathbf{v}'\,\mathrm{d}t = \int \mathbf{F}\cdot\mathbf{v}\,\mathrm{d}t - \mathbf{u}\cdot\int\mathbf{F}\,\mathrm{d}t = W - \mathbf{u}\cdot\mathbf{J}

Why work is frame-dependent at all: the same force acts over different displacements in different frames — a force applied to a body that is at rest in your frame does no work for you, but plenty for an observer who sees the body moving.

Step 4. Compare Steps 2 and 3: both $\Delta K$ and $W$ shift by exactly $-\mathbf{u}\cdot\mathbf{J}$ . So $W' = \Delta K'$ holds if and only if $W = \Delta K$ did — the work–energy theorem is Galilean covariant, even though neither side is invariant. The two-cart system provides a concrete check: boosting by $u = 3\ \mathrm{m/s}$ (into the CM frame) sends $K = 40.5\ \mathrm{J}$ to $40.5 - 3(18) + \tfrac12(6)(9) = 13.5\ \mathrm{J}$ , exactly the $K_{\mathrm{rel}}$ of König's split.

Limiting cases

case	result	meaning
body at rest, $v = 0$	$K = 0$	the floor of the account: kinetic energy is non-negative in every frame
low speed, $v \ll c$	$mc^2(\gamma - 1) \to \tfrac12 mv^2 + \tfrac38 mv^4/c^2 + \cdots$	the classical formula is the first term of the relativistic series; the correction reaches 1% at $v \approx 0.115c$
$v \to c$	$K \to \infty$	infinite work is needed to reach light speed — the deepest reason nothing with mass gets there
observer riding the CM	$K' = K_{\mathrm{rel}}$	the minimum kinetic energy any inertial observer can assign; pure internal motion
uniform circular motion	$K = \mathrm{const}$	force ⟂ velocity delivers zero power: acceleration without energy change
equal momenta, unequal masses	$K = p^2/2m$ : lighter body wins	at fixed $p$ , kinetic energy is inversely proportional to mass

Common trapTwice the speed is only twice as dangerous

This is the misconception the $v^2$ law exists to kill. Kinetic energy — and with it braking distance, crash energy, and the dent in 's Gravesande's clay — grows with the square of the speed: going from 50 to 100 km/h quadruples the energy your brakes must destroy, and a 30% speed increase already carries 69% more energy. The linear intuition ("a bit faster, a bit worse") systematically underestimates fast collisions, which is why speed limits, following the parabola, are far stricter than they feel. A related trap: kinetic energies of velocity components do add ( $K = \tfrac12 m v_x^2 + \tfrac12 m v_y^2$ ) — but kinetic energies of different bodies moving oppositely never cancel, because $K$ is built from a square and cannot be negative.

Worked examples

Example 1 — the braking parabola (confidence builder). A 1200 kg car brakes hard on dry asphalt ( $\mu = 0.8$ ), first from 50 km/h, then from 100 km/h. Find the kinetic energy and stopping distance in each case, taking $g = 9.81\ \mathrm{m/s^2}$ .

Step 1. Convert speeds: $50\ \mathrm{km/h} = 13.89\ \mathrm{m/s}$ and $100\ \mathrm{km/h} = 27.78\ \mathrm{m/s}$ .

Step 2. Kinetic energies from (1): $K_{50} = \tfrac12(1200)(13.89)^2 = 1.16\times10^{5}\ \mathrm{J} \approx 116\ \mathrm{kJ}$ , and $K_{100} = \tfrac12(1200)(27.78)^2 = 4.63\times10^{5}\ \mathrm{J} \approx 463\ \mathrm{kJ}$ — four times as much, as $v^2$ demands.

Step 3. The friction force available is $F = \mu m g = 0.8(1200)(9.81) = 9418\ \mathrm{N}$ , roughly constant during the skid.

Step 4. Apply the work–energy theorem (4) with $W = -Fd$ draining the account to zero: $d = K/F = mv^2/(2\mu m g) = v^2/(2\mu g)$ . The mass cancels — a loaded truck and a hatchback skid the same distance at the same speed and the same $\mu$ .

Step 5. Numbers: $d_{50} = (13.89)^2/(2\cdot0.8\cdot9.81) = \mathbf{12.3\ m}$ and $d_{100} = (27.78)^2/(2\cdot0.8\cdot9.81) = \mathbf{49.2\ m}$ — exactly a factor of 4, ten car lengths instead of two and a half.

Example 2 — two carts and the energy that collisions can spend (exam level). A 2 kg cart moving at 6 m/s catches up with a 4 kg cart moving at 1.5 m/s on a frictionless track. Using König's theorem, find how much kinetic energy is lost if they couple together, and check the elastic alternative.

Step 1. Total momentum and CM velocity: $p = 2(6) + 4(1.5) = 18\ \mathrm{kg\,m/s}$ , so $v_{\mathrm{cm}} = 18/6 = 3\ \mathrm{m/s}$ . This is frozen by momentum conservation no matter what happens in the collision.

Step 2. Total kinetic energy before: $K = \tfrac12(2)(6)^2 + \tfrac12(4)(1.5)^2 = 36 + 4.5 = 40.5\ \mathrm{J}$ .

Step 3. König split (7): the untouchable CM part is $K_{\mathrm{cm}} = \tfrac12(6)(3)^2 = 27\ \mathrm{J}$ ; the spendable relative part is $K_{\mathrm{rel}} = 40.5 - 27 = 13.5\ \mathrm{J}$ . (Direct check: relative to the CM the carts move at $+3$ and $-1.5\ \mathrm{m/s}$ , giving $\tfrac12(2)(9) + \tfrac12(4)(2.25) = 9 + 4.5 = 13.5\ \mathrm{J}$ .)

Step 4. Perfectly inelastic outcome: the coupled 6 kg pair moves at $v_{\mathrm{cm}} = 3\ \mathrm{m/s}$ with $K_f = 27\ \mathrm{J}$ . Energy lost: $\mathbf{13.5\ J}$ — precisely all of $K_{\mathrm{rel}}$ , converted to heat and deformation in the coupling. No collision consistent with momentum conservation could lose more.

Step 5. Elastic alternative: the standard 1D elastic formulas give $v_1' = \frac{(m_1 - m_2)v_1 + 2m_2 v_2}{m_1+m_2} = \frac{(-2)(6) + 8(1.5)}{6} = \mathbf{0}$ and $v_2' = \frac{(m_2-m_1)v_2 + 2m_1 v_1}{6} = \frac{3 + 24}{6} = \mathbf{4.5\ m/s}$ . The light cart stops dead. Check the books: $K_f = 0 + \tfrac12(4)(4.5)^2 = 40.5\ \mathrm{J}$ (conserved, as elastic requires) and $p_f = 4(4.5) = 18\ \mathrm{kg\,m/s}$ ✓. In the elastic case $K_{\mathrm{rel}}$ is not spent but redistributed — the relative speed reverses from $4.5$ to $-4.5\ \mathrm{m/s}$ .

Origins · 1 minGottfried Wilhelm Leibniz and how this was discovered

Gottfried Wilhelm Leibniz

1646–1716 · Leipzig and Hanover

Leibniz — philosopher, diplomat, co-inventor of the calculus — lit the fuse of the energy concept in 1686 with a short paper accusing Descartes of confusing two different measures of motion. The Cartesians held that the universe conserves "quantity of motion," $m|v|$ ; Leibniz showed with falling bodies that the quantity nature actually trades against height is $mv^2$ , which he named vis viva, the living force. The ensuing dispute ran for half a century, partly because both sides were right about different things — momentum and energy are each conserved, and answer different questions. Émilie du Châtelet's Institutions de physique (1740) fused Leibniz's dynamics with 's Gravesande's clay-ball experiments and, through her translation of Newton, carried vis viva into the mainstream; the modern factor of ½ and the name "kinetic energy" came only in the nineteenth century.

History

1686

— Leibniz's Brevis Demonstratio attacks Descartes' mv, proposing vis viva mv² as the true measure of motion

1722

— Willem 's Gravesande drops brass balls into clay: dent depth scales with v², experimental teeth for vis viva

1740

— Émilie du Châtelet's Institutions de physique synthesizes Leibniz's dynamics with the experiments and settles the dispute's terms

1829

— Gaspard-Gustave de Coriolis defines work as force times distance and fixes the modern factor: kinetic energy is ½mv², not mv²

1849

— William Thomson (Lord Kelvin) coins the phrase kinetic energy, completing the modern vocabulary

Going further

Texts. Morin, Introduction to Classical Mechanics, ch. 5 (Energy) · Taylor, Classical Mechanics, ch. 4 · Goldstein, Poole & Safko, Classical Mechanics, 3rd ed., ch. 1 · Landau & Lifshitz, Mechanics, §§4, 8 · Kleppner & Kolenkow, An Introduction to Mechanics, ch. 5. Key results to keep in mind: the binomial expansion of $\gamma$ has coefficients $1, \tfrac12, \tfrac38$ ; the König cross-term cancels because $\sum m_i\mathbf{v}_i' = 0$ ; the Galilean boost law is $K' = K - \mathbf{u}\cdot\mathbf{P} + \tfrac12 Mu^2$ with work and $\Delta K$ sharing the same shift $-\mathbf{u}\cdot\mathbf{J}$ . The worked numbers throughout — braking distances $12.3$ m and $49.2$ m ( $v^2/2\mu g$ with $\mu = 0.8$ ), the cart split $40.5 = 27.0 + 13.5$ J with elastic outcome $v_1' = 0$ , $v_2' = 4.5$ m/s, the $1\%$ relativistic threshold $v \approx 0.115c$ , DART's $1.1\times10^{10}$ J, and the EV's $0.19$ kWh — are all illustrative arithmetic derived from the boxed formulas.

Builds on: Velocity and Acceleration · Newton's Second Law · Vectors, Coordinate Systems and Kinematics · Relative Motion and Galilean Transformations. Leads to: Circular Motion and Centripetal Acceleration · Gravitational Force Near Earth's Surface.

Where does the ½ in ½mv² actually come from?

From calculus, not from physics. The work–energy derivation integrates $\mathbf{v}\cdot\mathrm{d}\mathbf{v}$ , and integrating anything against its own differential produces a half: $\int v\,\mathrm{d}v = \tfrac12 v^2$ , exactly like $\int x\,\mathrm{d}x = \tfrac12 x^2$ . Historically the half arrived late — Leibniz's vis viva was $mv^2$ , without it — and was fixed by Coriolis in 1829 so that kinetic energy would equal the work done in producing the motion, with no stray factor of 2. Nothing about the physics changes if you track $mv^2$ instead; every equation just sprouts compensating factors. The convention that survives is the one that makes $W_{\mathrm{net}} = \Delta K$ read cleanly.

If kinetic energy is frame-dependent, is energy conservation frame-dependent too?

Conservation holds in every inertial frame — what changes between frames is the conserved total, not the fact of conservation. The boost derivation shows why: under $\mathbf{v}' = \mathbf{v} - \mathbf{u}$ , both the kinetic-energy change and the work done shift by the identical amount $-\mathbf{u}\cdot\mathbf{J}$ , so any ledger that balanced before still balances. In a collision, momentum conservation guarantees the frame-dependent pieces cancel between the bodies: if kinetic energy is conserved in one frame (elastic), it is conserved in all frames. Observers disagree about how much energy there is — never about whether the books close.

Why do we need both momentum and kinetic energy? They look like the same information

They answer different questions about the same motion. Momentum $m\mathbf{v}$ is a vector tracking force × time (impulse); kinetic energy $\tfrac12 mv^2$ is a scalar tracking force × distance (work). A truck and a bullet can have equal momentum with wildly different energies, and two identical cars moving oppositely have zero total momentum but plenty of total kinetic energy. In collisions the split is decisive: momentum is conserved always, kinetic energy only when the collision is elastic — the gap between the two is exactly the energy dissipated. The 60-year Leibniz–Descartes dispute happened because both sides held one true half of this picture.

A body in uniform circular motion is accelerating the whole time. How can its kinetic energy be constant?

Because kinetic energy responds to power, not to acceleration. The rate of change is $\mathrm{d}K/\mathrm{d}t = \mathbf{F}\cdot\mathbf{v}$ , and a centripetal force points at the center — perpendicular to the velocity — so the dot product vanishes at every instant. The force continuously changes the direction of $\mathbf{v}$ while leaving its magnitude alone, and $K$ cares only about the magnitude. This is a whole family of physics in one observation: string tension in a whirled stone, the normal force on a bead in a bowl, the magnetic force $q\mathbf{v}\times\mathbf{B}$ , and the Coriolis force all steer without ever charging the energy account.

Where does the kinetic energy actually go in a car crash?

Into microscopic degrees of freedom, with König's theorem dictating the amount. Momentum conservation fixes the center-of-mass motion, so the crash can only spend the relative kinetic energy $K_{\mathrm{rel}}$ — in a head-on collision between equal cars, that is all of it. The spending happens through plastic deformation: crumple zones are engineered to fold at large, roughly constant force over half a metre or so, converting ordered motion into the disordered vibration of metal atoms — heat — plus sound and fracture energy. A crashed car is measurably warmer. None of the energy is destroyed; it is demoted from the one coherent mode (bulk motion) into $10^{24}$ incoherent ones, which is why the process cannot run backward.

Can kinetic energy ever be negative?

Not in classical mechanics — $K = \tfrac12 mv^2$ is a mass (positive) times a square (non-negative), so $K \geq 0$ in every inertial frame, with zero only for a body at rest in that frame. What can be negative are changes ( $\Delta K < 0$ when a body slows) and other energy accounts: potential energy has an arbitrary zero and is often negative (a bound satellite has $E = K + U < 0$ overall). When you meet a genuinely negative kinetic-energy-like quantity — as in quantum tunneling, where $E < V$ makes the classical $K$ formally negative inside a barrier — it is a flag that the particle picture has broken down, not that a classical body is moving with imaginary speed.

At what point does ½mv² stop being accurate, and what replaces it?

The exact expression is relativistic: $K = mc^2(\gamma - 1)$ with $\gamma = (1 - v^2/c^2)^{-1/2}$ . Expand it for small $v/c$ and the first term is exactly $\tfrac12 mv^2$ , with the leading correction $\tfrac38 mv^4/c^2$ . The correction reaches 1% of the classical value when $\tfrac34 v^2/c^2 = 0.01$ , i.e. $v \approx 0.115c \approx 35{,}000\ \mathrm{km/s}$ — far beyond any macroscopic motion (a spacecraft at 11 km/s is off by parts in $10^{9}$ ). Electrons cross the threshold at a few kilovolts, which is why accelerator physics and electron microscopy are relativistic trades, while planetary dynamics happily stays classical.

Doesn't the p²/2m form contradict ½mv² for light particles?

They are the same statement, rearranged: substitute $p = mv$ into $p^2/2m$ and $\tfrac12 mv^2$ falls out. The two forms feel contradictory because they hold different things fixed. At fixed speed, kinetic energy grows with mass ( $K = \tfrac12 mv^2$ : a truck beats a scooter). At fixed momentum, kinetic energy shrinks with mass ( $K = p^2/2m$ : the recoiling gun carries far less energy than its bullet). Neither statement is more fundamental; which one you reach for depends on which variable your problem conserves or specifies. Collisions conserve momentum, so $p^2/2m$ is the natural collision currency — and it is the form that survives into quantum mechanics, where momentum, not velocity, labels the states.

Kinetic Energy

Where this lives today

The physical picture

The energy of motion, one formula at a time

Systems of particles: the König split

Changing frames without breaking the books

Limiting cases

Worked examples

Going further

Before this

Where this leads