Fracture mechanics — stress intensity factor K, toughness K_IC

Traditional strength-of-materials assumes flawless parts. Reality? Manufacturing leaves micro-cracks. Meteoroid impacts create new ones. Temperature cycling fatigues joints. Fracture mechanics asks: "Given this crack exists, will it kill us?"

What Is the Stress Intensity Factor K?

Units: $\text{MPa}\sqrt{\text{m}}$ or $\text{psi}\sqrt{\text{in}}$

Why does K exist? Near a sharp crack, stress becomes theoretically infinite (stress concentration). But the shape of the stress field around any crack tip follows a universal mathematical form. K is the single number that scales universal pattern to match your specific crack.

Deriving K from First Principles

Start with a crack of length $2a$ in an infinite plate under remote tensile stress $\sigma$.

Step 1: Stress concentration at crack tip At distance $r$ from the crack tip, angle $\theta$ from the crack plane, elasticity theory (solving Airy stress functions with crack boundary conditions) gives:

$\sigma_{y}(r, \theta) = \frac{K_I}{\sqrt{2\pi r}} \cos\frac{\theta}{2}\left(1 - \sin\frac{\theta}{2}\sin\frac{3\theta}{2}\right) + \text{higher order terms}$

Why this form? The $1/\sqrt{r}$ singularity comes from the crack tip being a mathematical sharp notch — stress must blow up as $r \to 0$ to satisfy force equilibrium at a point-like stress concentrator. The $\cos(\theta/2)$ angular dependence comes from satisfying traction-free crack faces (no stress perpendicular to the crack surfaces). Note the minus sign in the bracket $[1 - \sin(\theta/2)\sin(3\theta/2)]$ — this is the correct Westergaard/Williams result; along the crack plane ($\theta = 0$) both stress components reduce to $K_I/\sqrt{2\pi r}$.

Step 2: Defining $K_I$ For a Mode I crack (opening mode, tensile), comparing the leading-order stress field to remote loading:

$K_I = \sigma \sqrt{\pi a} \cdot Y$

where:

• $\sigma$ = remote applied stress
• $a$ = crack length (for edge crack) or half-length (for center crack)
• $Y$ = geometry correction factor (depends on crack shape, boundary conditions)

Why $\sqrt{\pi a}$? Dimensional analysis: $K$ has units $\text{stress} \times \sqrt{\text{length}}$. The $\pi$ emerges from integrating the stress field around the elliptical crack front in the exact solution. For an infinite plate with center crack: $Y = 1$.

Why this step? This connects the abstract stress field parameter $K$ to measurable engineering quantities: applied stress and crack size.

Solution: $K_I = \sigma \sqrt{\pi a} \cdot Y$

Substitute: $K_I = 150 \times 10^6 \times \sqrt{\pi \times 0.002} \times 1.12$

Calculate step-by-step:

• $\pi a = 3.1416 \times 0.002 = 0.006283$ m
• $\sqrt{0.006283} = 0.07927$ m$^{1/2}$
• $K_I = 150 \times 10^6 \times 0.07927 \times 1.12 = 13.31 \times 10^6$ Pa·m$^{1/2}$

$K_I = 13.3 \text{ MPa}\sqrt{\text{m}}$

Why this step? Converting to standard units lets us compare to tabulated material toughness values.

Solution: For center crack, use half-length $a = 4$ mm = 0.004 m: $K_I = \sigma \sqrt{\pi a} \times 1.0 = 80 \times 10^6 \times \sqrt{\pi \times 0.004}$

Calculate step-by-step:

• $\pi a = 3.1416 \times 0.004 = 0.012566$ m
• $\sqrt{0.012566} = 0.1121$ m$^{1/2}$
• $K_I = 80 \times 10^6 \times 0.1121 = 8.97 \times 10^6$ Pa·m$^{1/2}$

$K_I = 8.97 \text{ MPa}\sqrt{\text{m}}$

Why calculate this? To assess whether the panel survives deployment loads — we'll compare to $K_{IC}$ next.

What Is Fracture Toughness $K_{IC}$?

The subscript "I" = Mode I (opening), "C" = critical value.

Physical meaning: $K_{IC}$ quantifies a material's resistance to crack growth. High $K_{IC}$ = tough material (crack needs large stress to propagate). Low $K_{IC}$ = brittle material (crack propagates easily).

Failure Criterion

A structure fails by fracture when:

$K_I \geq K_{IC}$

Why? When the applied stress intensity $K_I$ reaches the material's toughness limit $K_{IC}$, the crack-tip stress field exceeds the material's cohesive strength at the atomic scale — bonds break faster than plastic deformation can blunt the crack, causing rapid propagation.

$K_I \leq \frac{K_{IC}}{n}$

Rearranging for maximum allowable crack size:

$a_{\text{max}} = \frac{1}{\pi} \left( \frac{K_{IC}}{Y \sigma n} \right)^2$

Derivation: Start with $K_I = \sigma \sqrt{\pi a} Y$, set $K_I = K_{IC}/n$:

$\sigma \sqrt{\pi a} Y = \frac{K_{IC}}{n}$

Solve for $a$: $\sqrt{\pi a} = \frac{K_{IC}}{Y \sigma n}$

$a = \frac{1}{\pi} \left( \frac{K_{IC}}{Y \sigma n} \right)^2$

Why this step? This is the damage-tolerance design equation — it tells inspectors the minimum detectable crack size needed to ensure safety.

Solution: $a_{\text{max}} = \frac{1}{\pi} \left( \frac{35}{1.12 \times 120 \times 2} \right)^2$

Calculate numerator: $\frac{35}{1.12 \times 120 \times 2} = \frac{35}{268.8} = 0.1302 \text{ m}^{1/2}$

Square: $(0.1302)^2 = 0.01695$

Divide by $\pi$: $a_{\text{max}} = \frac{0.01695}{3.1416} = 0.00539 \text{ m} = 5.4 \text{ mm}$

Interpretation: Non-destructive inspection must detect cracks smaller than 5.4 mm. Eddy current or ultrasonic methods typically detect 1-2 mm cracks, providing safety margin.

Why this matters: This determines inspection intervals — if fatigue grows cracks at 0.5 mm/year, inspect every 2 years to catch cracks before critical size.

Three Modes of Fracture

Most spacecraft failures are Mode I. Mixed-mode fracture uses $K_{\text{eff}} = \sqrt{K_I^2 + K_{II}^2 + K_{III}^2}$ compared to $K_{IC}$.

Representative Toughness Values

Why care? Material selection balances strength (high $\sigma_y$), toughness ($K_{IC}$), and density. High-strength alloys often have lower toughness — fracture mechanics prevents over-optimization.

Why it feels right: We're taught stress < yield strength = safe.

What's wrong: A 10 mm crack in that titanium under 200 MPa gives $K_I = 200 \times \sqrt{\pi \times 0.01} \times 1.0 = 35.4$ MPa$\sqrt{\text{m}}$. If $K_{IC} = 55$ MPa$\sqrt{\text{m}}$, it's safe. But if the titanium was improperly heat-treated and $K_{IC}$ dropped to 30 MPa$\sqrt{\text{m}}$, the part fractures despite stress being far below yield.

The fix: Always check both strength (bulk stress < yield) and toughness (crack-tip $K < K_{IC}$). Fracture mechanics governs when cracks are present.

Why it feels right: The formula looks simple.

What's wrong: For a corner crack at a hole, $Y = 3.5$. Your $K$ estimate is 3.5× too low — you think it's safe when it's not.

The fix: Look up $Y$ for your actual geometry in handbooks (Tada, Paris, Irwin; Rooke & Cartwright). Finite element models compute $Y$ for complex shapes.

Or: "Krazy Inspectors Catch cracks" → $K_I$ = critical value is $K_{IC}$ when crack propagates.

$K$ is like a "danger number" that measures how much that tiny spot is being pulled. $K_{IC}$ is the "breaking number" for the rubber — if the danger number gets bigger than the breaking number, rip!

Engineers check: does our spaceship have any cuts? How big? How hard are we pulling? If the danger number stays below the breaking number, the spaceship won't rip apart in space.

Connections to Other Concepts

• Stress concentration factors — $K$ is the fracture-specific version for cracks; $K_t$ is for holes/notches in elastic regime
• Griffith energy criterion — $K_{IC}$ derived from energy balance. For plane stress: $K_{IC} = \sqrt{2 E \gamma}$; for plane strain: $K_{IC} = \sqrt{\dfrac{2 E \gamma}{1 - \nu^2}}$ (Young's modulus $E$, surface energy $\gamma$, Poisson's ratio $\nu$)
• Fatigue crack growth (Paris law) — $da/dN = C (\Delta K)^m$, uses $\Delta K = K_{\max} - K_{\min}$ to predict crack growth per cycle
• J-integral — energy-based fracture parameter for elastic-plastic materials; $J = K^2/E$ in linear-elastic limit
• Non-destructive testing (NDT) — eddy current, ultrasonic, radiography detect cracks before $a$ reaches $a_{\text{critical}}$
• Damage tolerance philosophy — design assumes cracks exist, size inspection intervals to catch growth

Active Recall Questions

#flashcards/physics

Concept Map

Hinglish (regional understanding)