🔥 JEE Main 2027 Session 1 is in January — start the JEE Mains Physics Test Series →

CLASS 11 · MATHEMATICAL TOOLS

Differentiation in Physics

Differentiate position once and you get velocity, differentiate again and you get acceleration, again and you get jerk. The same single operation, done over and over, is also how you find the launch angle that throws a projectile farthest, or the position where a spring's stored energy sits lowest: just set the derivative to zero and solve.

01 · See it

Watch it happen

Projectile range vs. launch angle, fixed speed v = 20 m/s

35.35 m
Range R(θ)
0.712 m/deg
dR/dθ (tangent slope)
Still rising
Where you are

Drag the slider toward 45° and watch the tangent (orange) flatten until it turns green at dR/dθ = 0, exactly the optimum. Keep dragging past it and the slope goes negative, confirming this zero is a maximum and not just a flat spot the curve passes through.

This is a position-independent curve, the launch angle on the x-axis against the resulting range on the y-axis, for a fixed launch speed. Drag the angle slider and watch the tangent line’s slope, that’s dR/dθ, shrink toward zero as you approach 45°, then swing negative past it. The same “drag a parameter, watch its derivative cross zero at the optimum” idea is exactly what you’d do for a potential energy curve U(x), just with position on the x-axis instead of angle.

02 · Derive it

Where it comes from

The chain of derivatives. Start with position x(t)x(t). Differentiating once gives the rate at which position changes, velocity:

Velocity
v(t)=dxdtv(t) = \dfrac{dx}{dt}

Differentiate again (the rate at which velocity itself changes) and you get acceleration:

Acceleration
a(t)=dvdt=d2xdt2a(t) = \dfrac{dv}{dt} = \dfrac{d^2x}{dt^2}

Nothing stops you from differentiating a third time, the rate at which acceleration changes is called jerk:

Jerk
j(t)=dadt=d3xdt3j(t) = \dfrac{da}{dt} = \dfrac{d^3x}{dt^3}

Each arrow in this chain is the exact same operation (take the slope of the previous quantity’s graph against time), just applied one more time.

Instantaneous vs. average. The average velocity over an interval from t1t_1 to t2t_2 is just the secant slope on a position-time graph:

vavg=x(t2)−x(t1)t2−t1v_{\text{avg}} = \dfrac{x(t_2)-x(t_1)}{t_2-t_1}

Shrinking that interval to zero width turns the secant into the tangent at a single instant, exactly the derivative, exactly the idea the “Reading Physics Graphs” page builds slope and area from:

Instantaneous velocity, as a limit
v(t)=lim⁡Δt→0x(t+Δt)−x(t)Δt=dxdtv(t) = \lim_{\Delta t \to 0} \dfrac{x(t+\Delta t)-x(t)}{\Delta t} = \dfrac{dx}{dt}

The same limit, applied to v(t)v(t) instead of x(t)x(t), gives instantaneous acceleration. A position-time graph’s tangent line at any point literally is the instantaneous velocity there, drawn as a slope.

Maximizing projectile range. A projectile launched at speed vv and angle θ\theta above level ground has horizontal and vertical velocity components vcos⁡θv\cos\theta and vsin⁡θv\sin\theta. Time of flight (up and back down to the same height) is T=2vsin⁡θgT = \dfrac{2v\sin\theta}{g}, so the range is horizontal speed times that time:

R(θ)=(vcos⁡θ)⋅T=2v2sin⁡θcos⁡θg=v2sin⁡(2θ)gR(\theta) = (v\cos\theta)\cdot T = \dfrac{2v^2\sin\theta\cos\theta}{g} = \dfrac{v^2\sin(2\theta)}{g}

To find the angle that maximizes RR, differentiate with respect to θ\theta and set the result to zero:

dRdθ=2v2cos⁡(2θ)g\dfrac{dR}{d\theta} = \dfrac{2v^2\cos(2\theta)}{g}
Setting dR/dθ = 0
2v2cos⁡(2θ)g=0  ⇒  cos⁡(2θ)=0  ⇒  2θ=90∘  ⇒  θ=45∘\dfrac{2v^2\cos(2\theta)}{g} = 0 \;\Rightarrow\; \cos(2\theta) = 0 \;\Rightarrow\; 2\theta = 90^\circ \;\Rightarrow\; \theta = 45^\circ

A zero derivative alone only says the curve is flat there, not that it’s a maximum, so check the second derivative:

d2Rdθ2=−4v2sin⁡(2θ)g∣θ=45∘=−4v2g<0\dfrac{d^2R}{d\theta^2} = -\dfrac{4v^2\sin(2\theta)}{g}\bigg|_{\theta=45^\circ} = -\dfrac{4v^2}{g} < 0

Negative confirms a maximum, so θ=45∘\theta = 45^\circ really does give the largest possible range, with Rmax⁡=v2/gR_{\max} = v^2/g.

Minimizing potential energy. The same move, differentiate and set to zero, locates equilibrium positions on any potential energy curve U(x)U(x), since the force is F=−dUdxF = -\dfrac{dU}{dx} and equilibrium means zero net force. Take a toy interatomic potential with a repulsive and an attractive term:

A model potential
U(x)=Ax12−Bx6U(x) = \dfrac{A}{x^{12}} - \dfrac{B}{x^{6}}

Differentiate and set the slope to zero to find the equilibrium separation x0x_0:

dUdx=−12Ax13+6Bx7=0\dfrac{dU}{dx} = -\dfrac{12A}{x^{13}} + \dfrac{6B}{x^{7}} = 06Bx7=12Ax13  ⇒  x6=2AB  ⇒  x0=(2AB)1/6\dfrac{6B}{x^{7}} = \dfrac{12A}{x^{13}} \;\Rightarrow\; x^{6} = \dfrac{2A}{B} \;\Rightarrow\; x_0 = \left(\dfrac{2A}{B}\right)^{1/6}

Checking d2U/dx2>0d^2U/dx^2 > 0 at x0x_0 confirms it’s a minimum, not a maximum, exactly the stable separation at which two atoms sit (the bottom of the potential well), matching how a spring’s own U(x)=12kx2U(x) = \tfrac{1}{2}kx^2 has its one minimum at x=0x=0 by the identical logic, just with simpler algebra.

03 · Break it

Where the shortcut stops working

Swapping average rate for instantaneous rate is the single most common slip. A car that covers 100 km in 2 hours has an average speed of 50 km/h, but that tells you nothing about its speed at any one instant, it could have been stopped at a light for ten minutes and doing 90 km/h the rest of the way. Only dx/dtdx/dt evaluated at a specific tt gives the instantaneous value; dividing a finite displacement by a finite time never does, no matter how short that interval looks.

Finding where a derivative is zero and immediately calling it “the maximum” skips a real step. The projectile range function does have its one interior zero of dR/dθdR/d\theta at 45°, and the second derivative test confirms it’s a maximum there, but that confirmation is required, not optional, the same zero-derivative condition is satisfied at minima and at inflection points too. Forgetting to check also means forgetting to compare against the domain’s endpoints: here R(0∘)=R(90∘)=0R(0^\circ) = R(90^\circ) = 0, both clearly worse than the interior point, but that’s a check, not an assumption.

Treating acceleration as if it were always constant quietly throws away jerk. Constant-acceleration kinematics equations (the v=u+atv = u + at family) only apply while aa truly doesn’t change; a rocket burning fuel, a car easing off the accelerator, or a elevator ramping up to speed all have nonzero jerk, and plugging their changing acceleration into a constant-a formula gives an answer that looks precise and is simply wrong.

04 · Master it

Apply it under exam conditions

A ball is launched from level ground at speed v=22 m/sv = 22\text{ m/s}, with g=9.8 m/s2g = 9.8\text{ m/s}^2. Find (a) the launch angle that maximizes its range, confirming with the second derivative that it truly is a maximum, and (b) the value of that maximum range.

(a) The range as a function of angle is:

R(θ)=v2sin⁡(2θ)gR(\theta) = \dfrac{v^2\sin(2\theta)}{g}

Differentiate with respect to θ and set it to zero:

dRdθ=2v2cos⁡(2θ)g=0  ⇒  cos⁡(2θ)=0  ⇒  θ=45∘\dfrac{dR}{d\theta} = \dfrac{2v^2\cos(2\theta)}{g} = 0 \;\Rightarrow\; \cos(2\theta)=0 \;\Rightarrow\; \theta = \textcolor{#e08a1e}{45^\circ}

Check the second derivative at this angle:

d2Rdθ2∣45∘=−4v2sin⁡(90∘)g=−4(22)29.8=−197.6\dfrac{d^2R}{d\theta^2}\bigg|_{45^\circ} = -\dfrac{4v^2\sin(90^\circ)}{g} = -\dfrac{4(22)^2}{9.8} = \textcolor{#e08a1e}{-197.6}

Negative, so θ = 45° is confirmed a maximum, not merely a flat point.

(b) Substitute θ = 45° back into R(θ), where sin(90°) = 1:

Rmax⁡=v2sin⁡(90∘)g=(22)29.8=4849.8=49.4 mR_{\max} = \dfrac{v^2\sin(90^\circ)}{g} = \dfrac{(22)^2}{9.8} = \dfrac{484}{9.8} = \textcolor{#e08a1e}{49.4\text{ m}}

Cross-check: evaluating R(θ)R(\theta) directly at 30° and 60° both give about 42.8 m, symmetric about 45° and clearly smaller than 49.4 m, exactly the shape a single interior maximum should produce.

05 · FAQs

Quick answers

Why is the maximum range at 45°, and does that change if launch and landing heights differ?+

For level ground, range is R(θ) = v²sin(2θ)/g, and sin(2θ) peaks at exactly 2θ = 90°, giving θ = 45°. That clean result depends entirely on the projectile landing at the same height it launched from. The instant the landing height differs (firing off a cliff, or onto a raised platform), the time of flight is no longer symmetric in θ, the range formula changes, and the optimal angle shifts away from 45°, lower when landing below launch height, higher when landing above it.

What is jerk, and why does it ever matter?+

Jerk is the derivative of acceleration, da/dt, the rate at which acceleration itself is changing. It matters whenever a changing acceleration is physically felt, not just computed: an elevator that snaps from rest straight into its full acceleration feels harsher than one that ramps up smoothly, even if both reach the same final acceleration, because the first has much larger jerk. Vehicle and rollercoaster design specifically limits jerk to keep rides comfortable, even when the velocity and acceleration limits alone would allow a sharper change.

If a derivative is zero, is that point always a maximum?+

No. A zero derivative only means the function is momentarily flat there, it could be a maximum, a minimum, or neither (an inflection point where the curve flattens but keeps rising, like y = x³ at x = 0). You have to check the second derivative's sign, or simply check both neighbors, before calling it a maximum or minimum.

What's the real difference between average and instantaneous velocity?+

Average velocity is total displacement divided by total time over some finite interval, exactly the secant slope between the interval's two endpoints on a position-time graph. Instantaneous velocity is the limit of that same ratio as the interval shrinks to zero, exactly the tangent slope at one single instant. They agree only when the motion is perfectly uniform over that interval; otherwise the average can hide large swings the instantaneous value reveals.

How does setting dU/dx = 0 locate equilibrium for a potential energy curve?+

Force is minus the slope of the potential energy curve, F = -dU/dx, so wherever that slope is zero, the force is zero too, meaning the system is in equilibrium there (no net push in either direction). Whether that equilibrium is stable (a minimum of U, like the bottom of a spring's potential well) or unstable (a maximum of U, like a ball balanced on a hilltop) again comes down to the second derivative's sign.