Linear Momentum and Impulse

Catch an egg with stiff hands and it shatters; catch it with hands that give, and it survives. Same egg, same speed, same change of motion — the only difference is time. An airbag, a gymnast's bent knees, a car's crumple zone, and a boxer rolling with a punch are all engineering the same trade: the change in motion is fixed, so stretch it over more milliseconds and the force shrinks in exact proportion. To make that trade precise you need the quantity Newton himself put at the center of mechanics — momentum, mass times velocity, his "quantity of motion." You already know the second law as $F = ma$ . Newton wrote it differently: force is the rate of change of momentum. That version is more honest — it survives when the mass itself changes, and it hands you the tool for collisions, impacts, and kicks the moment you integrate it over time. Nobody knows the force a bat exerts on a baseball; it screams from zero to tens of thousands of newtons and back in under a millisecond, following a jagged profile that differs from swing to swing. And yet the ball's change of velocity is perfectly predictable, because only the area under the force–time curve matters. That area is the impulse, and it equals the momentum change exactly, whatever the curve's shape. By the end you will compute the 18,000-newton handshake between bat and ball, then design a crumple zone — deriving why eighty centimeters of sacrificial steel turns a 255-kilonewton dashboard impact into a survivable 13-kilonewton ride.

Key takeaways

Momentum $\mathbf{p} = m\mathbf{v}$ is a vector "quantity of motion" — the time-integral of force needed to stop a body, direction included.
Newton's second law in its original form: $\mathbf{F} = \mathrm{d}\mathbf{p}/\mathrm{d}t$ ; the familiar $\mathbf{F} = m\mathbf{a}$ is its constant-mass special case.
Impulse is force integrated over time, and it buys momentum change: $\mathbf{J} = \int \mathbf{F}\,\mathrm{d}t = \Delta\mathbf{p}$ — the area under the force–time curve, exactly, whatever the curve's shape.
Same $\Delta\mathbf{p}$ , your choice of price: huge force for a moment, or gentle force for longer — the design equation of airbags is $\bar{\mathbf{F}} = \Delta\mathbf{p}/\Delta t$ .
Peaks depend on pulse shape: a triangular pulse peaks at twice the average force, a half-sine at $\pi/2 \approx 1.57$ times — and materials fail at the peak, not the average.
Summed over a system, internal forces cancel in third-law pairs: $\mathrm{d}\mathbf{P}/\mathrm{d}t = \mathbf{F}_{\mathrm{ext}}$ — the gateway to momentum conservation.
Key idea: force is the flow rate of momentum, $\mathbf{F} = \mathrm{d}\mathbf{p}/\mathrm{d}t$ — so a force changes momentum not by its strength or duration alone but by their accumulated product, the area under the force–time curve.

The core insight

Force is the time-rate of change of momentum, $\mathbf{F} = \mathrm{d}\mathbf{p}/\mathrm{d}t$ ; integrating over time gives the impulse–momentum theorem $\mathbf{J} = \int \mathbf{F}\,\mathrm{d}t = \Delta\mathbf{p}$ , and only that time-integral — not the force profile's shape — sets the change of motion.

Where this lives today

Crash-safety engineering is applied impulse–momentum, executed at industrial scale. Every Euro NCAP and NHTSA test wires the dummy with accelerometers and integrates acceleration–time histories; the Head Injury Criterion that decides a car's star rating is a weighted integral of exactly the curves in this article's figures. The design targets are stated in the theorem's language: keep the area fixed (physics), stretch the base (crumple depth, belt stretch, airbag venting), cap the peak (biomechanical limits). The same grammar governs climbing ropes — dynamic ropes are rated by the peak force they transmit while absorbing a standard fall's impulse — and sports engineering plays the game in reverse: "follow-through" in golf and tennis lengthens contact time to pump more impulse into the ball. At the other end of the scale, momentum bookkeeping is a discovery machine. Particle physicists at the LHC add up the transverse momenta of everything a collision produces; when the sum fails to balance, something invisible carried the difference away — the neutrino announced itself exactly this way in 1930, and "missing momentum" remains the signature dark-matter searches watch for in 2026. In between, every model-rocket motor is literally labeled by $\int F\,\mathrm{d}t$ — motor classes are brackets of total impulse in newton-seconds — and deep-space ion thrusters win on the same integral: micronewtons sustained for years out-accumulate a chemical engine's minutes of fury.

The physical picture

Momentum measures how hard something is to stop — with a memory of direction. A loaded truck rolling at walking pace and a rifle bullet can carry the same $mv$ ; what they share is the total force-times-time it takes to bring either to rest. That product is the natural currency of collisions, because in a real impact the force is a violent, millisecond-scale spike no instrument tracks cleanly. Mount a load cell on a wall and throw a ball at it: the readout's wiggles differ every throw, but integrate each spike over time and you get the same number to three digits — the ball's momentum change. The force curve is weather; its area is climate.

Figure

Two ways to pay the same impulse — the area is the deal

This one picture prices every safety device you own. The momentum change in a crash is non-negotiable: it is your mass times your speed. What the airbag negotiates is the shape of the force curve — the same area drawn low and wide instead of tall and narrow. A wine glass dropped on concrete and one dropped on a sofa suffer the same impulse; the concrete collects it in a tenth of a millisecond at shattering force, the cushion collects it over fifty times longer at a fiftieth of the peak. Nothing about the glass changed. Only the clock did. The second half of the picture is a bookkeeping miracle. When two bodies interact, Newton's third law says their forces on each other are equal and opposite at every instant — so their impulses are too, and whatever momentum one gains the other loses. Momentum is never created inside a system; it only changes hands. Watch two skaters push apart from rest: the lighter one leaves faster in exact inverse proportion to mass, and the pair's total momentum stays precisely zero. That observation, made systematic, becomes conservation of linear momentum — this article builds the tool, and the sequel article makes the conservation law its whole subject.

The math: momentum, the true second law, and impulse

Two definitions carry this whole article, and the first is Newton's choice of currency. Ask what makes a moving body hard to stop and neither mass nor velocity alone answers — the supertanker and the bullet are both formidable, for opposite reasons. The honest measure multiplies them:

\mathbf{p} = m\mathbf{v} \qquad (1)

where: $m$ — the (inertial) mass; $\mathbf{v}$ — the velocity; units $\mathrm{kg\,m/s}$ , a vector along the motion. The vector character is not decoration: two identical clay balls closing at $\pm 5\ \mathrm{m/s}$ carry zero total momentum, and that sign-sensitivity is precisely what Descartes missed and Huygens fixed (see the history band). Double either factor and you double the stopping job. With the currency defined, the second law says how force spends it. Newton's own statement in the Principia is not $m\mathbf{a}$ but "the change of motion is proportional to the motive force impressed" — force as the flow rate of momentum:

\mathbf{F}_{\mathrm{net}} = \frac{\mathrm{d}\mathbf{p}}{\mathrm{d}t} \qquad (2)

where: $\mathbf{F}_{\mathrm{net}}$ — the net force on the body. Expand the derivative by the product rule, $\mathrm{d}(m\mathbf{v})/\mathrm{d}t = \dot{m}\mathbf{v} + m\dot{\mathbf{v}}$ : for constant mass the first term dies and (2) hands back the familiar $m\mathbf{a}$ . For rockets and raindrops — bodies that shed or gather mass — equation (2) applied to the whole system is the form that stays honest, and it is the one that survives, with $\mathbf{p} = \gamma m \mathbf{v}$ , into relativity. Everything that follows is (2) processed two ways: integrate it over time and the impulse–momentum theorem appears; sum it over particles and the conservation law appears.

Which formula answers which question

you want	use	notes
how hard is it to stop	$\mathbf{p} = m\mathbf{v}$ , eq. (1)	a vector — signs and components matter
force from a known momentum flow	$\mathbf{F} = \mathrm{d}\mathbf{p}/\mathrm{d}t$ , eq. (2)	the true second law; $m\mathbf{a}$ only if $\dot m = 0$
effect of a messy impact	$\mathbf{J} = \int \mathbf{F}\,\mathrm{d}t = \Delta\mathbf{p}$ , eq. (3)	shape of $F(t)$ irrelevant
estimate a collision force	$\bar{\mathbf{F}} = \Delta\mathbf{p}/\Delta t$ , eq. (4)	multiply by 1.6–2 for the damaging peak
peak from pulse shape	eqs. (5)–(6)	triangle ×2, half-sine ×1.57
many bodies at once	$\mathrm{d}\mathbf{P}/\mathrm{d}t = \mathbf{F}_{\mathrm{ext}}$ , eq. (7)	internal forces cancel pairwise

Derivation AThe impulse-momentum theorem

We are deriving the central result of this article, equation (3): the time-integral of the net force equals the change in momentum, exactly and regardless of the force profile's shape. The route is the shortest in mechanics — integrate the second law (2) over the duration of the encounter and let the fundamental theorem of calculus do the work. The result is exact for any force history, which is the entire point: in a real collision $F(t)$ is a jagged, unrepeatable spike, and the theorem tells us we never need to know it. What to watch is the vector character — the theorem holds component by component, and a rebound (velocity changing sign) changes momentum by more than a dead stop does.

Step 1. Multiply (2) by $\mathrm{d}t$ and integrate over the interaction, from first touch $t_1$ to separation $t_2$ :

\int_{t_1}^{t_2} \mathbf{F}_{\mathrm{net}}\,\mathrm{d}t = \int_{t_1}^{t_2} \frac{\mathrm{d}\mathbf{p}}{\mathrm{d}t}\,\mathrm{d}t

Why: the instantaneous force during a millisecond impact is unknowable, but its time-integral is pinned by the endpoints alone.

Step 2. The right side telescopes by the fundamental theorem of calculus into $\mathbf{p}(t_2) - \mathbf{p}(t_1)$ . Why: integrating a rate of change recovers the total change — the interior history cancels out, which is exactly why the pulse's shape is irrelevant to $\Delta\mathbf{p}$ .

Step 3. Name the left side the impulse $\mathbf{J}$ and conclude:

\mathbf{J} = \int_{t_1}^{t_2} \mathbf{F}\,\mathrm{d}t = \Delta\mathbf{p} \qquad (3)

where: $\mathbf{J}$ — impulse, in $\mathrm{N\,s} = \mathrm{kg\,m/s}$ ; geometrically, the area under the force–time curve, component by component. The matching units are the structural tell: impulse is measured in the same currency as momentum because it is delivered momentum.

Step 4. Divide by the duration to define the average force — the constant force that would deliver the same area in the same window:

\bar{\mathbf{F}}\,\Delta t = \Delta\mathbf{p} \qquad (4)

where: $\Delta t = t_2 - t_1$ — the contact time. Why this is the design equation of every crumple zone: the right side is fixed by the collision, so force and time trade off in exact inverse proportion — stretch $\Delta t$ , shrink $\bar F$ . *Verified: for a constant force, (3) reduces to $\mathbf{F}\,\Delta t = m\mathbf{v}_f - m\mathbf{v}_i$ , matching constant-acceleration kinematics; Taylor ch. 3, Kleppner and Kolenkow ch. 4.*

The average force is a summary, not the whole story, and its blind spot has teeth: materials and skulls fail at the peak force, not the average. The peak depends on the shape of the pulse, so an engineer needs conversion factors between the rectangle of equation (4) and the shapes real impacts draw.

Go deeper · Derivation B · 1 minPeak versus average force for standard pulse shapes

Derivation BPeak versus average force for standard pulse shapes

The goal is a set of shape factors connecting the easy quantity — the average force $\bar F = J/\Delta t$ of equation (4) — to the damaging one, the peak $F_{\mathrm{peak}}$ , for the three canonical force–time profiles: rectangle, triangle, and half-sine. The strategy is pure area bookkeeping: each shape must enclose the same impulse $J$ in the same window $\Delta t$ , so its peak is fixed by its geometry. The half-sine is the case to watch — it requires one actual integral, and it is the shape real elastic impacts approximate best.

Step 1. Rectangular pulse: $F$ holds constant for $\Delta t$ , so the curve is its own average and $F_{\mathrm{peak}} = \bar F$ . Why: this is the gentlest possible delivery of a given area in a given time — the theoretical floor that crumple-zone designers chase.

Step 2. Triangular pulse: $F$ ramps linearly to $F_{\mathrm{peak}}$ at $\Delta t/2$ , then back down. A triangle's area is half its bounding rectangle's, so $J = \tfrac12 F_{\mathrm{peak}}\,\Delta t$ , giving

F_{\mathrm{peak}} = \frac{2J}{\Delta t} = 2\,\bar F \qquad (5)

Why the factor 2: holding the same area on a shrinking summit means the summit must rise — halve the effective width, double the height.

Step 3. Half-sine pulse, $F(t) = F_{\mathrm{peak}}\sin(\pi t/\Delta t)$ : the area is $J = F_{\mathrm{peak}} \int_0^{\Delta t} \sin(\pi t/\Delta t)\,\mathrm{d}t = \tfrac{2}{\pi} F_{\mathrm{peak}}\,\Delta t$ , so

F_{\mathrm{peak}} = \frac{\pi}{2}\,\bar F \approx 1.57\,\bar F \qquad (6)

Why: the sine integral evaluates to $2\Delta t/\pi$ — substitute $u = \pi t/\Delta t$ and the antiderivative $-\cos u$ picks up 2 across the half-cycle. Real measured impacts typically land between the half-sine's 1.6 and the triangle's 2. *Verified: recomputing the integral, $\int_0^{\Delta t}\sin(\pi t/\Delta t)\,\mathrm{d}t = [-(\Delta t/\pi)\cos(\pi t/\Delta t)]_0^{\Delta t} = (\Delta t/\pi)(1+1) = 2\Delta t/\pi$ — the shape factors match the standard impact-engineering values.*

One more consequence of (3) costs a line and pays forever. A ball arriving at speed $v$ and stopping needs $|\Delta p| = mv$ ; a ball arriving at $v$ and rebounding at $e v$ (restitution coefficient $e$ ) needs $|\Delta p| = mv(1+e)$ — up to twice as much for a perfectly elastic bounce, because the wall must first cancel the incoming momentum and then supply outgoing momentum in the opposite direction. Reversal costs double. For a $0.06\ \mathrm{kg}$ tennis ball at $30\ \mathrm{m/s}$ and a 5 ms contact, the full rebound demands $\bar F = 2(0.06)(30)/0.005 = 720\ \mathrm{N}$ — over a thousand times the ball's weight, which is the quantitative license to ignore gravity during the hit. Finally, sum the second law over many particles instead of integrating it over time, and the third law turns the crank:

Go deeper · Derivation C · 1 minThe system law - internal forces cancel

Derivation CThe system law - internal forces cancel

We are deriving equation (7), the momentum law for a whole system of interacting particles: the total momentum responds to external forces only, however violent the internal ones. The strategy is to sum equation (2) over every particle, split each force into external and internal parts, and watch Newton's third law annihilate the internal sum pair by pair. This three-line argument is the gateway to conservation of momentum — the sequel article builds its entire collision toolkit on it — so the step to watch is the pairing: the cancellation needs forces equal, opposite, and summed over both members of each pair.

Step 1. For particles $i = 1, 2, \dots$ with total momentum $\mathbf{P} = \sum_i m_i \mathbf{v}_i$ , sum (2):

\frac{\mathrm{d}\mathbf{P}}{\mathrm{d}t} = \sum_i \mathbf{F}_i^{\mathrm{ext}} + \sum_{i \neq j} \mathbf{F}_{ij}

Why: each particle feels the outside world plus its neighbors; $\mathbf{F}_{ij}$ is the force on $i$ from $j$ .

Step 2. Group the internal forces into action–reaction pairs: $\mathbf{F}_{ij} + \mathbf{F}_{ji} = \mathbf{0}$ for every pair, by Newton's third law. Why the whole sum dies: it is a sum of such pairs, each identically zero — the internal drama, however violent, is momentum-neutral.

Step 3. What survives is the system law:

\frac{\mathrm{d}\mathbf{P}}{\mathrm{d}t} = \mathbf{F}_{\mathrm{ext}}, \qquad \mathbf{P} = \sum_i m_i \mathbf{v}_i = M\mathbf{v}_{\mathrm{cm}} \qquad (7)

where: $M$ — the total mass; $\mathbf{v}_{\mathrm{cm}}$ — the velocity of the center of mass. The center of mass obeys a single-particle second law, deaf to all internal shoving. Set $\mathbf{F}_{\mathrm{ext}} = \mathbf{0}$ and $\mathbf{P}$ is constant — each Cartesian component separately, since (7) is a vector equation. Verified: Goldstein §1.2 and Landau-Lifshitz §7, where the deeper origin is exhibited — invariance under spatial translation via Noether's theorem, which is why the conservation law outlives the third law itself in field theory.

Limiting cases

Case	Result	Meaning
constant mass	$\mathbf{F} = \mathrm{d}(m\mathbf{v})/\mathrm{d}t = m\mathbf{a}$	the familiar second law — the special case, not the fundamental one
constant force	$J = F\,\Delta t$	the area rule reduces to base times height — the rectangle
impulsive limit $\Delta t \to 0$ at fixed $J$	$F_{\mathrm{peak}} \to \infty$ , $mg\,\Delta t \to 0$	idealized "impulsive force": finite $\Delta p$ , unbounded force, gravity invisible
dead stop vs elastic rebound	$\|J\| = mv$ vs $\|J\| = 2mv$	reversal costs double — the wall stops the ball, then throws it back
long, gentle push	same $\Delta p$ from tiny $F$	an ion thruster's millinewtons, integrated over months, rival a chemical rocket's roar
$v \to c$	$\mathbf{p} = \gamma m\mathbf{v}$	the definition upgrades; $\mathbf{F} = \mathrm{d}\mathbf{p}/\mathrm{d}t$ and the theorem survive into relativity

Common trap“The heavier object hits harder”

Force, momentum, and impulse are three different quantities, and this misconception fuses them. Compare a $1000\ \mathrm{kg}$ car rolling at $2\ \mathrm{m/s}$ into a wall, stopping over $1\ \mathrm{s}$ , with a $0.05\ \mathrm{kg}$ bullet at $400\ \mathrm{m/s}$ embedding in $10^{-4}\ \mathrm{s}$ . The car carries $2000\ \mathrm{kg\,m/s}$ — one hundred times the bullet's $20\ \mathrm{kg\,m/s}$ . Yet the average forces are $2000\ \mathrm{N}$ for the car and $200{,}000\ \mathrm{N}$ for the bullet: the "lighter" impact delivers a hundred times the force. Momentum sets the area under the force–time curve; the contact time sets the height. Never infer force from momentum without asking how long the delivery took. And two companion traps ride along: momentum is a vector (a rebound changes $\mathbf{p}$ by more than a stop), and momentum is not kinetic energy — the vector $m\mathbf{v}$ survives every isolated collision while the scalar $\tfrac12 mv^2$ survives only elastic ones.

Worked examples

1 · The two catches (confidence builder)

A $0.15\ \mathrm{kg}$ baseball arrives at $40\ \mathrm{m/s}$ and is caught. Find the average force on the hands if the catch takes (a) $5\ \mathrm{ms}$ — rigid hands — or (b) $50\ \mathrm{ms}$ — gloved hands riding backward with the ball.

Step 1. The momentum change is the same either way: $|\Delta p| = 0.15 \times 40 = 6.0\ \mathrm{kg\,m/s}$ , so by equation (3) the impulse the hands must deliver is $6.0\ \mathrm{N\,s}$ , whatever they do.

Step 2. Rigid catch, equation (4): $\bar F = 6.0/0.005 = \boxed{1200\ \mathrm{N}}$ — about the weight of a large adult, concentrated on your palm.

Step 3. Soft catch: $\bar F = 6.0/0.050 = \boxed{120\ \mathrm{N}}$ — ten times the time, one tenth the force. Same purchase, different installment plan.

Step 4. Sanity-check the neglect of gravity: its impulse over 50 ms is $mg\,\Delta t = 0.15 \times 9.8 \times 0.05 = 0.074\ \mathrm{N\,s}$ — about 1% of the catch impulse, invisible in the accounting, exactly as the impulsive limit promises.

2 · Designing the crumple zone (exam level)

A $1400\ \mathrm{kg}$ car carrying a $75\ \mathrm{kg}$ driver hits a rigid barrier at $17\ \mathrm{m/s}$ (about $61\ \mathrm{km/h}$ ). The crumple zone collapses over $d = 0.80\ \mathrm{m}$ at roughly constant force. Compare the force on a belted, airbag-protected driver with that on an unbelted driver who hits the dashboard and stops in $5\ \mathrm{ms}$ .

Step 1. Get the crash duration from the crumple depth. With roughly constant deceleration, $v^2 = 2ad$ gives $a = 17^2/(2 \times 0.80) = 181\ \mathrm{m/s^2} \approx 18\,g$ , and the stopping time is $\Delta t = v/a = 17/181 = 0.094\ \mathrm{s}$ .

Step 2. The car's own ledger: $\Delta p_{\mathrm{car}} = 1400 \times 17 = 2.38 \times 10^4\ \mathrm{N\,s}$ , so the barrier pushes with $\bar F = 2.38 \times 10^4 / 0.094 \approx \boxed{2.5 \times 10^5\ \mathrm{N}}$ — a quarter of a meganewton, absorbed by sacrificial steel.

Step 3. Belted driver, riding the crumple down: $\Delta p = 75 \times 17 = 1275\ \mathrm{N\,s}$ delivered over the full $0.094\ \mathrm{s}$ : $\bar F = 1275/0.094 = \boxed{13.6\ \mathrm{kN}}$ — the same $18\,g$ as the car. Punishing but survivable, because belt and airbag spread it across the torso's strongest structures.

Step 4. Unbelted driver: the car stops, but the driver keeps moving at $17\ \mathrm{m/s}$ until the dashboard collects the same $1275\ \mathrm{N\,s}$ in $0.005\ \mathrm{s}$ : $\bar F = 1275/0.005 = \boxed{255\ \mathrm{kN}}$ — about $350\,g$ , roughly nineteen times the belted case and far beyond survivable limits.

Step 5. Read the design principle from the ratio: the impulse was identical in Steps 3 and 4 — a fixed area under $F(t)$ . All the engineering — $0.80\ \mathrm{m}$ of collapsing steel, a belt that pre-tensions, an airbag inflating in $30\ \mathrm{ms}$ — exists to stretch the base of that area from 5 milliseconds to 94, collapsing the height by the same factor.

Figure

Same area, three shapes — and the crash engineer's version of the choice

Origins · 1 minIsaac Newton and how this was discovered

Isaac Newton

1643–1727 · Woolsthorpe, Cambridge, London

Newton's second law, as he actually wrote it in the Principia (1687), is a momentum statement: "the change of motion is proportional to the motive force impressed" — his motus, quantity of motion, being exactly $m\mathbf{v}$ . He was building on a generation of collision science: Descartes had proposed that God conserves the total "quantity of motion" but reckoned it as an unsigned scalar, which fails; Huygens, Wallis, and Wren, answering a Royal Society challenge in 1668, got the collision rules right by treating motion as directed. Newton absorbed their results, grounded them in his third law, and handed physics its first exact conservation principle. Characteristically, he stated the law in the impulse-friendly form — change of motion, not rate — because his geometric proofs in the Principia proceed by discrete blows and limiting arguments rather than by explicit derivatives. It took Euler, in 1750, to recast the law as the differential equation $\mathbf{F} = \mathrm{d}\mathbf{p}/\mathrm{d}t$ per coordinate; and it took Emmy Noether, in 1918, to show that the conservation law Newton grounded in action–reaction actually flows from the homogeneity of space itself.

History

1644

— Descartes' Principia Philosophiae proposes conservation of "quantity of motion", reckoned as an unsigned scalar — the right instinct crippled by the missing sign.

1668

— Wallis, Wren, and Huygens, answering a Royal Society challenge, deliver correct collision laws by treating momentum as a directed quantity; the books finally balance.

1687

— Newton's Principia states the second law as the change of motion produced by the impressed force, and grounds momentum conservation in the third law.

1750

— Euler's "Discovery of a new principle of mechanics" recasts the law as differential equations, making the impulse-momentum theorem a routine consequence of calculus.

1959

— Barenyi's crumple zone (patented 1952) reaches production in the Mercedes W111: the car body becomes an impulse-stretching machine.

Going further

Texts. Kleppner & Kolenkow, An Introduction to Mechanics, Ch. 4 (momentum, impulsive forces) · Morin, Introduction to Classical Mechanics, Ch. 4–5 · Taylor, Classical Mechanics, Ch. 3 · Goldstein, Poole & Safko, Classical Mechanics, 3rd ed., §1.1–1.2 · Landau & Lifshitz, Mechanics, §7–9 for the symmetry viewpoint.

Provenance. All equations are standard textbook results. A verifier should: confirm $\mathbf{F}=\mathrm{d}\mathbf{p}/\mathrm{d}t\to m\mathbf{a}$ via the product rule (Goldstein §1.1); recompute the impulse–momentum theorem by integrating the second law (Taylor Ch. 3); rederive two-body conservation from Newton's third law (Morin Ch. 5); rederive the Tsiolkovsky equation $\Delta v=u\ln(m_0/m_f)$ (Goldstein, Taylor Ch. 3); and rederive the 1D elastic-collision velocity formulas (Morin Ch. 5). Numerical check examples use only arithmetic with given values. Worked-example numbers on this page (catch forces $1200\ \mathrm{N}$ and $120\ \mathrm{N}$ ; crumple-zone crash $\Delta t = 0.094\ \mathrm{s}$ , $13.6\ \mathrm{kN}$ belted vs $255\ \mathrm{kN}$ unbelted; the half-sine shape factor $\pi/2$ ; tennis-ball rebound $\bar F = 720\ \mathrm{N}$ ) are illustrative arithmetic recomputed from the boxed formulas.

Builds on: Velocity and Acceleration · Newton's Second Law · Vectors in Kinematics and Coordinate Systems · Newton's First Law and Inertial Frames. Leads to: Newton's Third Law · Relative Motion and Galilean Transformations · Circular Motion and Centripetal Acceleration.

Are momentum and kinetic energy basically the same thing?

No — they are not even the same kind of thing. Momentum $m\mathbf{v}$ is a vector, linear in speed; kinetic energy $\tfrac12 mv^2$ is a scalar, quadratic in speed. The distinction has teeth: momentum is conserved in every collision of an isolated system, kinetic energy only in elastic ones. Two identical clay balls at $\pm 5\ \mathrm{m/s}$ carry zero total momentum but $25\ \mathrm{J}$ of kinetic energy — they stick, stop dead, and momentum is perfectly conserved while every joule vanishes into heat and deformation. Useful cross-check: $K = p^2/2m$ , so at fixed momentum, lighter objects carry more kinetic energy.

Why is

\mathbf{F} = \mathrm{d}\mathbf{p}/\mathrm{d}t

called more fundamental than

F = ma

Because it remains true in three situations where $m\mathbf{a}$ fails or misleads. First, variable mass: a rocket or an accreting raindrop exchanges momentum with the mass it sheds or gathers, and only the momentum form applied to the whole system books that correctly. Second, relativity: with $\mathbf{p} = \gamma m\mathbf{v}$ , the law $\mathbf{F} = \mathrm{d}\mathbf{p}/\mathrm{d}t$ survives verbatim while $\mathbf{F} = m\mathbf{a}$ does not (force and acceleration are not even parallel in general). Third, it is Newton's own statement — the Principia's Law II is about change of motion, i.e. of momentum. The product rule shows $m\mathbf{a}$ is the $\dot m = 0$ special case.

Does the impulse depend on the shape of the force curve at all?

The impulse — and therefore the momentum change — depends only on the area, never the shape. That is the content of the theorem: integrating $\mathrm{d}\mathbf{p}/\mathrm{d}t$ telescopes to the endpoints. What the shape controls is the peak force, and that is where damage lives. For the same area in the same window, a rectangle peaks at $\bar F$ , a half-sine at $1.57\,\bar F$ , a triangle at $2\,\bar F$ (equations 5–6). So two impacts with identical impulse can differ by a factor of two in what they break. Engineers negotiate the shape precisely because physics has already fixed the area.

If a huge truck and a bullet can have the same momentum, why does one crush and the other pierce?

Because force is momentum per unit time, and the two deliver their momentum on wildly different clocks. The truck's $2000\ \mathrm{kg\,m/s}$ arriving over a second is a couple of kilonewtons — a shove. A bullet's $20\ \mathrm{kg\,m/s}$ arriving over a tenth of a millisecond is $200\ \mathrm{kN}$ concentrated on a few square millimeters — enormous force and enormous pressure. Momentum tells you the size of the ledger entry; contact time and contact area decide whether the delivery feels like a push, a punch, or a puncture.

Why exactly is it legitimate to ignore gravity during a collision?

It is a quantitative approximation, not a convention. Every force present contributes its own impulse, $\int F\,\mathrm{d}t$ . During a bat–ball contact of $0.7\ \mathrm{ms}$ , gravity contributes $mg\,\Delta t = 1.4 \times 7 \times 10^{-4} \approx 10^{-3}\ \mathrm{N\,s}$ , while the contact force delivers about $13\ \mathrm{N\,s}$ — ten thousand times more. As $\Delta t \to 0$ , any force that stays finite contributes vanishing impulse; only the contact spike, which grows as $\Delta t$ shrinks, survives the accounting. Over the seconds of flight that follow, the situation reverses: gravity's impulse accumulates and dominates. Always compare $mg\,\Delta t$ to $\Delta p$ before dropping it.

The wall a ball bounces off doesn't move. Did momentum conservation fail?

No — the wall is bolted to the Earth, and the pair momentum change $2mv$ went into the planet. With $M_\oplus \approx 6 \times 10^{24}\ \mathrm{kg}$ , the Earth's recoil velocity from a tennis ball is around $10^{-24}\ \mathrm{m/s}$ — real, exactly what the third law demands, and utterly unmeasurable. This is the general pattern whenever one partner is enormous: momentum conservation holds perfectly, but the big partner's share hides in a microscopic velocity. It is also why we may treat "ball hits rigid wall" as a fixed-boundary problem while still trusting the impulse bookkeeping on the ball alone.

Is momentum the same in every reference frame?

No. Under a Galilean boost to a frame moving at velocity $\mathbf{V}$ , each particle's momentum shifts: $\mathbf{p}' = \mathbf{p} - m\mathbf{V}$ . A parked truck has zero momentum to you and plenty to a passing train. What is frame-independent (among inertial frames) is the law: momentum changes are produced by impulses in every frame, and if total momentum is conserved in one inertial frame it is conserved in all of them, because the shift $M\mathbf{V}$ is constant in time. Choosing the frame where $\mathbf{P} = 0$ — the center-of-mass frame — is the single most useful trick in collision analysis.

My textbook says impulse J = F·Δt, but yours has an integral. Which is right?

Both — $F\,\Delta t$ is the constant-force special case, and more usefully it is the definition of the average force when $F$ varies: $\bar F \equiv J/\Delta t$ , the height of the rectangle with the same area as the true pulse (the dashed rectangle in the hero figure). For estimates this is exactly what you want, since collision durations are measurable while force profiles are not. Just remember its blind spot: the average says nothing about the peak, and the peak is what breaks bones — multiply by 1.6–2 for realistic pulse shapes when safety is the question.