Laws of change — equations that describe how systems evolve
Uranium-238 decays to thorium-234, which decays to protactinium-234:
$^{238}\text{U} \xrightarrow{k_1} {}^{234}\text{Th} \xrightarrow{k_2} {}^{234}\text{Pa}$
The amount of thorium depends on:
$$\frac{d[\text{Th}]}{dt} = k_1[\text{U}] - k_2[\text{Th}]$$
This is a differential equation — an equation relating a function to its derivatives.
Solving it tells us how [Th] changes over time.
Differential equations combine our primitives:
CHANGE (dy/dt) + RATE (how fast) + ACCUMULATION (integration to solve)
A DE is a statement about rates. Solving it reveals the function.
| Property | Description |
|---|---|
| Order | Highest derivative (1st, 2nd, etc.) |
| Linearity | y and derivatives to first power only |
| ODE vs PDE | One vs multiple independent variables |
$$\frac{d[A]}{dt} = -k[A]$$
Solution: $[A] = [A]_0 e^{-kt}$
$$\frac{d[A]}{[A]} = -k \, dt$$
$$\int \frac{d[A]}{[A]} = -k \int dt$$
$$\ln[A] = -kt + C \quad \Rightarrow \quad [A] = [A]_0 e^{-kt}$$
B is produced from A and consumed to form C:
$$\frac{d[B]}{dt} = k_1[A] - k_2[B]$$
$[A] = [A]_0 e^{-k_1 t}$
$[B] = \frac{k_1[A]_0}{k_2 - k_1}\left(e^{-k_1 t} - e^{-k_2 t}\right)$
$[C] = [A]_0 - [A] - [B]$
$$ay'' + by' + cy = 0$$
Method: Try $y = e^{rx}$ → Characteristic equation: $ar^2 + br + c = 0$
$$y = C_1 e^{r_1 x} + C_2 e^{r_2 x}$$
$$y = (C_1 + C_2 x)e^{rx}$$
$$y = e^{\alpha x}(C_1 \cos\beta x + C_2 \sin\beta x)$$
This gives oscillatory solutions!
$$m\frac{d^2x}{dt^2} + \gamma\frac{dx}{dt} + kx = 0$$
Divide by m: $x'' + 2\beta x' + \omega_0^2 x = 0$
| Condition | Regime | Behavior |
|---|---|---|
| β < ω₀ | Underdamped | Oscillates with decay |
| β = ω₀ | Critical | Fastest decay, no oscillation |
| β > ω₀ | Overdamped | Slow exponential decay |
Time-independent Schrödinger equation with V = 0 inside:
$$-\frac{\hbar^2}{2m}\frac{d^2\psi}{dx^2} = E\psi$$
$$\psi_n(x) = \sqrt{\frac{2}{L}}\sin\left(\frac{n\pi x}{L}\right)$$
$$E_n = \frac{n^2 h^2}{8mL^2}$$
Boundary conditions + DE → quantized energy!
When analytical solutions fail, step forward numerically:
$$y_{n+1} = y_n + h \cdot f(x_n, y_n)$$
For serious work: Use Runge-Kutta (RK4) or adaptive methods (Julia's DifferentialEquations.jl, Python's scipy.integrate).
| Type | Form | Method |
|---|---|---|
| Separable | dy/dx = g(x)h(y) | Separate, integrate |
| Linear | dy/dx + Py = Q | Integrating factor μ = e^(∫Pdx) |
| Roots | Solution Form |
|---|---|
| r₁, r₂ distinct real | C₁e^(r₁x) + C₂e^(r₂x) |
| r repeated | (C₁ + C₂x)e^(rx) |
| α ± βi complex | e^(αx)(C₁cos βx + C₂sin βx) |
| System | Equation | Solution |
|---|---|---|
| 1st-order kinetics | d[A]/dt = -k[A] | [A]₀e^(-kt) |
| 2nd-order kinetics | d[A]/dt = -k[A]² | [A]₀/(1+kt[A]₀) |
| Harmonic oscillator | x'' + ω²x = 0 | A cos(ωt + φ) |