Study guide
5.2 Static Failure Theories for Brittle Materials & Fracture Mechanics
PE Mechanical: Machine Design and Materials Exam
Track your progress as you read
Finished sections get a check mark.
1 Chapter 1: Basic Engineering Practice: Drawings, Economics & Design Methodology
1.1 Engineering Terms, Symbols, Drawing Interpretation & Conventions 1.2 Project Management & Engineering Economics 1.3 Design Methodology, Risk Assessment & Safety
2 Chapter 2: Physical & Mechanical Properties of Engineering Materials
2.1 Stress-Strain Relationships & Fundamental Mechanical Properties 2.2 Physical Properties, Temperature Effects, Hardness, Impact Toughness & Creep
3 Chapter 3: Mechanics of Materials: Fundamental Loadings & Deflections
3.1 Axial Loading, Normal Stress & Statically Indeterminate Members 3.2 Direct Shear, Bearing & Transverse Shear in Beams 3.3 Torsion of Shafts, Power Transmission & Thin-Walled Tubes 3.4 Beam Bending Stresses & Deflection Analysis
4 Chapter 4: Mechanics of Materials: Stability, Multiaxial Stress & Thermal Effects
4.1 Column Buckling & Elastic/Inelastic Instability 4.2 Combined Loading, Principal Stresses & Mohr's Circle 4.3 Thermal Stresses & Interference Stresses
5 Chapter 5: Failure Theories: Static Yielding & Fracture Mechanics
5.1 Static Failure Theories for Ductile Materials: von Mises & Tresca 5.2 Static Failure Theories for Brittle Materials & Fracture Mechanics
6 Chapter 6: Fatigue Analysis & Dynamic Failure
6.1 High-Cycle Fatigue, S-N Curves & Marin Modifying Factors 6.2 Fluctuating Stresses, Failure Criteria & Cumulative Damage
7 Chapter 7: Mechanical Attachments: Fasteners & Bolted Joints
7.1 Threaded Fasteners, Proof Strength & Preload Mechanics 7.2 Bolted Joint Stiffness, External Loading & Separation 7.3 Non-Threaded Fasteners: Keys, Pins, Clevises, Retaining Rings, Rivets, Lugs & Anchors
8 Chapter 8: Mechanical Attachments: Permanent Joints & Bonding
8.1 Welded Joints & Weld Group Stress Analysis 8.2 Brazing, Soldering & Structural Adhesive Bonding
9 Chapter 9: Power Transmission: Shafts & Gearing
9.1 Transmission Shaft Design, Critical Speeds & Deflection 9.2 Spur Gears: Geometry, Kinematics & AGMA Stress Equations 9.3 Helical, Bevel & Worm Gearing
10 Chapter 10: Power Transmission: Bearings, Flexible Drives, Friction Devices & Prime Movers
10.1 Rolling-Element Bearings & L10 Life Calculations 10.2 Hydrodynamic Journal Bearings & Lubrication Regimes 10.3 Belts, Roller Chains & Flexible Drives 10.4 Clutches, Brakes & Power Screws 10.5 Electric Motors, Engines & Flywheels
11 Chapter 11: Mechanical Components: Pressure Vessels & Structural Systems
11.1 Pressure Vessels & Piping: Thin and Thick Walls, ASME VIII and B31 11.2 Mechanical Springs: Helical, Belleville & Leaf Springs 11.3 Structural Trusses, Rigid Frames & Energy Methods
12 Chapter 12: Mechanical Dynamics, Vibrations & Fluid Power
12.1 Single-Degree-of-Freedom Vibrations & Resonance 12.2 Vibration Isolation, Multi-DOF Systems & Transmissibility 12.3 Fluid Power Systems: Hydraulics & Pneumatics
13 Chapter 13: Kinematics, Mechatronics & Control Systems
13.1 Mechanism Kinematics, Linkages, Cams & Simple Machines 13.2 Basic Mechatronics, Sensors & Closed-Loop Controls
14 Chapter 14: Supportive Knowledge: Fits, Tolerances & Manufacturing Methods
14.1 Limits, Fits & Geometric Dimensioning & Tolerancing (GD&T) 14.2 Manufacturing Methods, Machinability, Heat Treatment & Assembly
15 Chapter 15: Supportive Knowledge: Codes, Computation, Testing & Corrosion
15.1 Engineering Codes & Standards 15.2 Computational Methods: FEA & Numerical Analysis 15.3 Instrumentation, Testing, Inspection & Quality 15.4 Chemical Processes: Corrosion, Oxidation & Embrittlement
5.2 Static Failure Theories for Brittle Materials & Fracture Mechanics
Key Takeaways Explain This Topic AI
Brittle materials exhibit percent elongation % E L < 5 % \%EL < 5\% % E L < 5% and significant strength asymmetry ( S u c ≫ S u t S_{uc} \gg S_{ut} S u c ≫ S u t , often 3 to 4 times) because internal micro-flaws open under tension but close under compression.
The Maximum Normal Stress (MNS) theory predicts fracture when σ 1 ≥ S u t \sigma_1 \ge S_{ut} σ 1 ≥ S u t or σ 3 ≤ − S u c \sigma_3 \le -S_{uc} σ 3 ≤ − S u c ; while valid in biaxial tension (Quadrant 1), it is dangerously non-conservative in Quadrant 4 where compressive stress accelerates tensile fracture.
The Modified Mohr (MM / Dowling) theory eliminates the unwarranted conservatism of the Brittle Coulomb-Mohr (BCM) theory in Quadrant 4 when ∣ σ B / σ A ∣ ≤ 1 |\sigma_B/\sigma_A| \le 1 ∣ σ B / σ A ∣ ≤ 1 ( n = S u t / σ A n = S_{ut}/\sigma_A n = S u t / σ A ), accurately tracking empirical biaxial test data.
Linear Elastic Fracture Mechanics (LEFM) governs crack propagation via the stress intensity factor K I = Y σ π a K_I = Y \sigma \sqrt{\pi a} K I = Yσ πa , predicting brittle fracture when K I ≥ K I c K_I \ge K_{Ic} K I ≥ K I c where K I c K_{Ic} K I c is the plane-strain toughness measured on ASTM E399 specimens sized by the 2.5 ( K I c / S y ) 2 2.5(K_{Ic}/S_y)^2 2.5 ( K I c / S y ) 2 rule.
The Leak-Before-Break (LBB) criterion requires that the critical crack size exceed the wall thickness ( a c > t a_c > t a c > t ), ensuring through-wall penetration produces detectable fluid leakage before catastrophic fast fracture can initiate.
Last updated: October 2026
5.2 Static Failure Theories for Brittle Materials & Fracture Mechanics
Quick Summary: Brittle materials sustain negligible plastic deformation prior to catastrophic fracture ( % E L < 5 % \%EL < 5\% % E L < 5% ) and exhibit markedly asymmetric strength, with ultimate compressive strengths exceeding tensile capacities by factors of three to four. Classical stress concentration formulations break down at sharp flaws where theoretical notch radii approach zero. This section covers the governing brittle failure criteria—Maximum Normal Stress, Brittle Coulomb-Mohr, and the Modified Mohr (Dowling) theory—and introduces Linear Elastic Fracture Mechanics (LEFM), including stress intensity ( K I K_I K I ), ASTM E399 plane strain fracture toughness ( K I c K_{Ic} K I c ), valid thickness constraints, NDE crack detection limits, and the Leak-Before-Break (LBB) design criterion.
1. Mechanical Behavior and Asymmetric Strength of Brittle Materials
Engineering materials exhibiting percent elongation at fracture % E L < 5 % \%EL < 5\% % E L < 5% are classified as brittle . Common engineering brittle materials include gray cast iron, white cast iron, hardened tool steels (quenched without tempering), structural ceramics (silicon nitride, alumina, zirconia), glass, concrete, and powder metallurgy components.
Cleavage Fracture vs. Plastic Slip
Unlike ductile metals that deform via dislocation glide along close-packed slip planes, brittle solids undergo cleavage fracture —the rapid, direct separation of atomic crystallographic planes normal to the maximum tensile stress direction. Because cleavage requires very little energy dissipation compared to plastic shear slip, brittle fracture is sudden, catastrophic, and unaccompanied by gross dimensional warning (no necking or yield plateau).
The Physical Origin of Compressive-to-Tensile Strength Asymmetry
The defining mechanical characteristic of brittle materials is their pronounced strength asymmetry :
S u c ≫ S u t S_{uc} \gg S_{ut} S u c ≫ S u t
For standard grades of gray cast iron (ASTM A48 Class 20 to Class 60), the ultimate compressive strength is typically three to four times greater than the ultimate tensile strength ( S u c ≈ 3 – 4 S u t S_{uc} \approx 3\text{--}4 S_{ut} S u c ≈ 3 – 4 S u t ):
Material Specification Tensile Strength S u t S_{ut} S u t Compressive Strength S u c S_{uc} S u c Ratio S u c / S u t S_{uc} / S_{ut} S u c / S u t
Gray Cast Iron Class 20 150 MPa 150\text{ MPa} 150 MPa ( 22 ksi 22\text{ ksi} 22 ksi ) 570 MPa 570\text{ MPa} 570 MPa ( 83 ksi 83\text{ ksi} 83 ksi ) 3.8 3.8 3.8
Gray Cast Iron Class 30 214 MPa 214\text{ MPa} 214 MPa ( 31 ksi 31\text{ ksi} 31 ksi ) 750 MPa 750\text{ MPa} 750 MPa ( 109 ksi 109\text{ ksi} 109 ksi ) 3.5 3.5 3.5
Gray Cast Iron Class 40 293 MPa 293\text{ MPa} 293 MPa ( 42.5 ksi 42.5\text{ ksi} 42.5 ksi ) 965 MPa 965\text{ MPa} 965 MPa ( 140 ksi 140\text{ ksi} 140 ksi ) 3.3 3.3 3.3
Structural Concrete 3 – 5 MPa 3\text{--}5\text{ MPa} 3 – 5 MPa ( 0.4 – 0.7 ksi 0.4\text{--}0.7\text{ ksi} 0.4 – 0.7 ksi ) 30 – 40 MPa 30\text{--}40\text{ MPa} 30 – 40 MPa ( 4.3 – 5.8 ksi 4.3\text{--}5.8\text{ ksi} 4.3 – 5.8 ksi ) 8 – 10 8\text{--}10 8 – 10
Sintered Alumina Ceramic 250 MPa 250\text{ MPa} 250 MPa ( 36 ksi 36\text{ ksi} 36 ksi ) 2500 MPa 2500\text{ MPa} 2500 MPa ( 360 ksi 360\text{ ksi} 360 ksi ) 10.0 10.0 10.0
This extreme asymmetry arises from internal microstructural defects. Gray cast iron contains an interconnected network of sharp, brittle graphite flakes dispersed throughout a pearlitic or ferritic iron matrix:
Under Tensile Stress: The sharp tip radii of graphite flakes act as severe internal stress concentrators. Tensile stresses open micro-cracks at flake tips, which link together and propagate across the cross section at low nominal tensile loads ( S u t S_{ut} S u t ).
Under Compressive Stress: Compressive forces push the surfaces of the graphite flakes and micro-cracks into mechanical contact. Flaws are squeezed shut, allowing the solid metallic iron matrix to transfer compressive load across the entire gross cross-sectional area. Failure can only occur when the iron matrix itself crushes or shears along oblique planes at very high stress levels ( S u c S_{uc} S u c ).
Fracture Morphology in Tension vs. Torsion
Because brittle fracture is governed by normal tensile cleavage:
Uniaxial Tension: Fractures along a flat plane oriented precisely perpendicular (90°) to the tensile loading axis.
Pure Torsion: Under pure shear stress τ \tau τ , Mohr's circle indicates that the principal stresses are σ 1 = + τ \sigma_1 = +\tau σ 1 = + τ and σ 3 = − τ \sigma_3 = -\tau σ 3 = − τ , acting on planes inclined at 45° to the shaft longitudinal axis. Consequently, a brittle cast iron shaft tested in pure torsion fractures along a characteristic 45° helical cleavage plane .
2. Maximum Normal Stress (MNS) Theory
Formulated by Scottish engineer William Rankine, the Maximum Normal Stress (MNS) theory is the simplest static failure model for brittle materials.
The MNS Postulate
Brittle fracture occurs when the maximum principal tensile stress in a multiaxial stress state reaches the ultimate tensile strength S u t S_{ut} S u t , or when the minimum principal stress reaches the ultimate compressive strength − S u c -S_{uc} − S u c .
Mathematical Formulation
Ordering principal stresses algebraically such that σ 1 ≥ σ 2 ≥ σ 3 \sigma_1 \ge \sigma_2 \ge \sigma_3 σ 1 ≥ σ 2 ≥ σ 3 , the MNS failure criterion states that fracture occurs if:
σ 1 ≥ S u t or σ 3 ≤ − S u c \sigma_1 \ge S_{ut} \quad \text{or} \quad \sigma_3 \le -S_{uc} σ 1 ≥ S u t or σ 3 ≤ − S u c
The design factor of safety ( n n n ) is evaluated separately for tension and compression:
n t = S u t σ 1 ( if σ 1 > 0 ) , n c = S u c ∣ σ 3 ∣ ( if σ 3 < 0 ) n_t = \frac{S_{ut}}{\sigma_1} \quad (\text{if } \sigma_1 > 0), \qquad n_c = \frac{S_{uc}}{|\sigma_3|} \quad (\text{if } \sigma_3 < 0) n t = σ 1 S u t ( if σ 1 > 0 ) , n c = ∣ σ 3 ∣ S u c ( if σ 3 < 0 )
The governing factor of safety is n = min ( n t , n c ) n = \min(n_t, n_c) n = min ( n t , n c ) .
Evaluation and Severe Limitations of MNS
In plane stress ( σ 3 = 0 \sigma_3 = 0 σ 3 = 0 ), the MNS criterion plots as a simple rectangle on the σ 1 − σ 2 \sigma_1 - \sigma_2 σ 1 − σ 2 plane bounded by σ 1 = S u t \sigma_1 = S_{ut} σ 1 = S u t , σ 1 = − S u c \sigma_1 = -S_{uc} σ 1 = − S u c , σ 2 = S u t \sigma_2 = S_{ut} σ 2 = S u t , and σ 2 = − S u c \sigma_2 = -S_{uc} σ 2 = − S u c :
Quadrant 1 (Biaxial Tension: σ 1 > 0 , σ 2 > 0 \sigma_1 > 0, \sigma_2 > 0 σ 1 > 0 , σ 2 > 0 ): MNS exhibits excellent agreement with experimental fracture data. Superimposed transverse tensile stress does not significantly alter the tensile fracture capacity.
Quadrant 4 (Combined Tension and Compression: σ 1 > 0 , σ 2 < 0 \sigma_1 > 0, \sigma_2 < 0 σ 1 > 0 , σ 2 < 0 ): Extensive experimental testing (notably by Coffin, Grassi, and Cornet on thin-walled gray iron tubes) reveals that MNS is dangerously non-conservative . When compressive stress is applied simultaneously with tensile stress, the tensile stress required to trigger brittle fracture drops well below S u t S_{ut} S u t . Designing with MNS in Quadrants 2 and 4 can lead to unexpected catastrophic rupture.
3. Coulomb-Mohr and Modified Mohr (MM) Theories
To account for strength asymmetry ( S u c > S u t S_{uc} > S_{ut} S u c > S u t ) and overcome the non-conservatism of MNS in mixed tension-compression stress states, two Mohr-based failure criteria are used in machine design.
Brittle Coulomb-Mohr (BCM) Theory
The Brittle Coulomb-Mohr (BCM) theory (also known as the internal friction theory) constructs a straight-line failure envelope tangent to the uniaxial tensile test circle (diameter S u t S_{ut} S u t ) and the uniaxial compressive test circle (diameter S u c S_{uc} S u c ) on a Mohr's stress plane.
For 2D plane stress with ordered in-plane principal stresses σ A ≥ σ B \sigma_A \ge \sigma_B σ A ≥ σ B :
Quadrant 1 ( σ A ≥ σ B ≥ 0 \sigma_A \ge \sigma_B \ge 0 σ A ≥ σ B ≥ 0 ): n = S u t σ A n = \frac{S_{ut}}{\sigma_A} n = σ A S u t
Quadrants 2 & 4 ( σ A ≥ 0 ≥ σ B \sigma_A \ge 0 \ge \sigma_B σ A ≥ 0 ≥ σ B ): σ A S u t − σ B S u c = 1 n ⟹ n = 1 σ A S u t − σ B S u c \frac{\sigma_A}{S_{ut}} - \frac{\sigma_B}{S_{uc}} = \frac{1}{n} \implies n = \frac{1}{\frac{\sigma_A}{S_{ut}} - \frac{\sigma_B}{S_{uc}}} S u t σ A − S u c σ B = n 1 ⟹ n = S u t σ A − S u c σ B 1
Quadrant 3 ( 0 ≥ σ A ≥ σ B 0 \ge \sigma_A \ge \sigma_B 0 ≥ σ A ≥ σ B ): n = − S u c σ B = S u c ∣ σ B ∣ n = -\frac{S_{uc}}{\sigma_B} = \frac{S_{uc}}{|\sigma_B|} n = − σ B S u c = ∣ σ B ∣ S u c
While BCM prevents the dangerous non-conservatism of MNS in Quadrant 4, extensive testing shows that BCM is excessively conservative when compressive stress is small ( ∣ σ B ∣ ≤ S u t |\sigma_B| \le S_{ut} ∣ σ B ∣ ≤ S u t ).
Modified Mohr (MM) Theory (Dowling Equations)
The Modified Mohr (MM) theory—standardized by N.E. Dowling and adopted in Shigley's Mechanical Engineering Design—modifies the BCM envelope to match experimental biaxial test data with exceptional precision. It recognizes that a minor compressive stress does not degrade the tensile fracture capacity until the compressive stress exceeds S u t S_{ut} S u t .
- For ordered in-plane principal stresses σ A ≥ σ B \sigma_A \ge \sigma_B σ A ≥ σ B : σB
- ^
- | Quadrant 1
Sut + - - - - - - - .
- | |
- | Modified |
- | Mohr |
- Suc | Envelope | Sut
- <-----------+--------------+---------------+--------+------> σA
- | | |
- | | |
- | | Quadrant 4 |
- | * - - - - - - - '
- | / (Sut, -Sut)
- | /
- | /
- | /
- Suc . - - - - '
- Quadrant 3
The factor of safety ( n n n ) is evaluated using Dowling's four distinct piecewise conditions:
Condition 1: Biaxial Tension (Quadrant 1: σ A ≥ σ B ≥ 0 \sigma_A \ge \sigma_B \ge 0 σ A ≥ σ B ≥ 0 )
Both principal stresses are tensile. Failure is governed strictly by tensile cleavage:
n = S u t σ A n = \frac{S_{ut}}{\sigma_A} n = σ A S u t
Condition 2: Tension-Dominant Mixed Stress (Quadrant 4: σ A ≥ 0 ≥ σ B \sigma_A \ge 0 \ge \sigma_B σ A ≥ 0 ≥ σ B with ∣ σ B / σ A ∣ ≤ 1 |\sigma_B / \sigma_A| \le 1 ∣ σ B / σ A ∣ ≤ 1 )
Here, the magnitude of the compressive stress is less than or equal to the tensile stress (meaning σ B ≥ − σ A \sigma_B \ge -\sigma_A σ B ≥ − σ A , which keeps the stress state above the − 45 ∘ -45^\circ − 4 5 ∘ diagonal line). Experimental data confirm that tensile fracture remains completely uninhibited by the transverse compression:
n = S u t σ A n = \frac{S_{ut}}{\sigma_A} n = σ A S u t
Condition 3: Compression-Dominant Mixed Stress (Quadrant 4: σ A ≥ 0 ≥ σ B \sigma_A \ge 0 \ge \sigma_B σ A ≥ 0 ≥ σ B with ∣ σ B / σ A ∣ > 1 |\sigma_B / \sigma_A| > 1 ∣ σ B / σ A ∣ > 1 )
When compressive stress exceeds tensile stress ( ∣ σ B ∣ > σ A |\sigma_B| > \sigma_A ∣ σ B ∣ > σ A ), the compressive stress significantly accelerates crack propagation. Dowling's linear interaction equation governs:
( S u c − S u t ) σ A S u c S u t − σ B S u c = 1 n \frac{(S_{uc} - S_{ut})\sigma_A}{S_{uc} S_{ut}} - \frac{\sigma_B}{S_{uc}} = \frac{1}{n} S u c S u t ( S u c − S u t ) σ A − S u c σ B = n 1
Solving explicitly for the factor of safety n n n :
n = S u c S u t ( S u c − S u t ) σ A − S u t σ B n = \frac{S_{uc} S_{ut}}{(S_{uc} - S_{ut})\sigma_A - S_{ut} \sigma_B} n = ( S u c − S u t ) σ A − S u t σ B S u c S u t
Condition 4: Biaxial Compression (Quadrant 3: 0 ≥ σ A ≥ σ B 0 \ge \sigma_A \ge \sigma_B 0 ≥ σ A ≥ σ B )
Both principal stresses are compressive. Failure is governed strictly by compressive crushing or shear splitting along the minimum principal stress axis:
n = − S u c σ B = S u c ∣ σ B ∣ n = -\frac{S_{uc}}{\sigma_B} = \frac{S_{uc}}{|\sigma_B|} n = − σ B S u c = ∣ σ B ∣ S u c
Quadrant-by-Quadrant Theory Comparison
Stress Quadrant Stress State Maximum Normal Stress (MNS) Brittle Coulomb-Mohr (BCM) Modified Mohr (MM)
Quadrant 1 σ A ≥ σ B ≥ 0 \sigma_A \ge \sigma_B \ge 0 σ A ≥ σ B ≥ 0 n = S u t σ A n = \frac{S_{ut}}{\sigma_A} n = σ A S u t n = S u t σ A n = \frac{S_{ut}}{\sigma_A} n = σ A S u t n = S u t σ A n = \frac{S_{ut}}{\sigma_A} n = σ A S u t
Quadrant 4 ( ∣ σ B / σ A ∣ ≤ 1 \lvert\sigma_B/\sigma_A\rvert \le 1 ∣ σ B / σ A ∣ ≤ 1 ) σ A > 0 > σ B \sigma_A > 0 > \sigma_B σ A > 0 > σ B n = S u t σ A n = \frac{S_{ut}}{\sigma_A} n = σ A S u t (Dangerously Non-conservative) n = 1 σ A S u t − σ B S u c n = \frac{1}{\frac{\sigma_A}{S_{ut}} - \frac{\sigma_B}{S_{uc}}} n = S u t σ A − S u c σ B 1 (Overly Conservative) n = S u t σ A n = \frac{S_{ut}}{\sigma_A} n = σ A S u t (Most Accurate)
Quadrant 4 ( ∣ σ B / σ A ∣ > 1 \lvert\sigma_B/\sigma_A\rvert > 1 ∣ σ B / σ A ∣ > 1 ) σ A > 0 > σ B \sigma_A > 0 > \sigma_B σ A > 0 > σ B n = min ( S u t σ A , S u c ∣ σ B ∣ ) n = \min(\frac{S_{ut}}{\sigma_A}, \frac{S_{uc}}{\lvert\sigma_B\rvert}) n = min ( σ A S u t , ∣ σ B ∣ S u c ) n = 1 σ A S u t − σ B S u c n = \frac{1}{\frac{\sigma_A}{S_{ut}} - \frac{\sigma_B}{S_{uc}}} n = S u t σ A − S u c σ B 1 n = S u c S u t ( S u c − S u t ) σ A − S u t σ B n = \frac{S_{uc} S_{ut}}{(S_{uc} - S_{ut})\sigma_A - S_{ut}\sigma_B} n = ( S u c − S u t ) σ A − S u t σ B S u c S u t
Quadrant 3 0 ≥ σ A ≥ σ B 0 \ge \sigma_A \ge \sigma_B 0 ≥ σ A ≥ σ B n = S u c ∣ σ B ∣ n = \frac{S_{uc}}{\lvert\sigma_B\rvert} n = ∣ σ B ∣ S u c n = S u c ∣ σ B ∣ n = \frac{S_{uc}}{\lvert\sigma_B\rvert} n = ∣ σ B ∣ S u c n = S u c ∣ σ B ∣ n = \frac{S_{uc}}{\lvert\sigma_B\rvert} n = ∣ σ B ∣ S u c
- 4. Linear Elastic Fracture Mechanics (LEFM)
- The Breakdown of Classical Stress Concentration
Classical elasticity analyzes geometric notches (fillets, holes, keyways) using stress concentration factors ( K t = σ max / σ nom K_t = \sigma_{\max} / \sigma_{\text{nom}} K t = σ m a x / σ nom ). For an elliptical hole with major axis 2 a 2a 2 a and notch root radius ρ \rho ρ oriented perpendicular to a tensile stress, the Inglis solution yields:
σ max = σ ( 1 + 2 a ρ ) \sigma_{\max} = \sigma\left(1 + 2\sqrt{\frac{a}{\rho}}\right) σ m a x = σ ( 1 + 2 ρ a )
In actual manufactured hardware, physical cracks, fatigue tears, and weld lack-of-fusion defects have atomic-scale root radii approaching zero ( ρ → 0 \rho \to 0 ρ → 0 ). Under classical elasticity, as ρ → 0 \rho \to 0 ρ → 0 , K t → ∞ K_t \to \infty K t → ∞ and σ max → ∞ \sigma_{\max} \to \infty σ m a x → ∞ . Infinite stress predictions make classical stress analysis incapable of assessing whether a cracked component can safely carry load. Linear Elastic Fracture Mechanics (LEFM) resolves this mathematical singularity by treating the crack tip not as an infinite point stress, but through energy dissipation and stress field intensity.
Griffith Energy Criterion (1921)
A.A. Griffith established that a crack propagates catastrophically when the rate of elastic strain energy released by infinitesimal crack advance ( d U / d a dU/da d U / d a ) equals or exceeds the rate of surface energy required to create two new fracture surfaces:
G = − d U d a ≥ G c = 2 γ s G = -\frac{dU}{da} \ge G_c = 2\gamma_s G = − d a d U ≥ G c = 2 γ s
For an ideal brittle material, the critical fracture stress is:
σ f = 2 E γ s π a \sigma_f = \sqrt{\frac{2E\gamma_s}{\pi a}} σ f = πa 2 E γ s
In engineering metals, E. Orowan and G.R. Irwin showed that localized dislocation plasticity always occurs in a microscopic zone at the crack tip, absorbing thousands of times more energy than surface tension alone ( γ p ≫ γ s \gamma_p \gg \gamma_s γ p ≫ γ s ). Irwin modified the criterion to:
G c = 2 ( γ s + γ p ) ≈ 2 γ p ⟹ σ f = E G c π a G_c = 2(\gamma_s + \gamma_p) \approx 2\gamma_p \implies \sigma_f = \sqrt{\frac{E G_c}{\pi a}} G c = 2 ( γ s + γ p ) ≈ 2 γ p ⟹ σ f = πa E G c
The Three Modes of Crack Surface Displacement
Crack tip stress fields are categorized into three fundamental kinematic modes:
Mode I (Opening Mode): Tensile normal stresses act perpendicular to the crack plane, pulling crack faces directly apart ( y y y -direction). Mode I is the predominant failure mode in mechanical machine design, pressure vessels, and rotating shafts.
Mode II (Sliding / In-Plane Shear Mode): Shear stresses act parallel to the crack plane and perpendicular to the crack front ( x x x -direction), causing crack faces to slide over one another.
Mode III (Tearing / Anti-Plane Shear Mode): Out-of-plane shear stresses act parallel to the crack front ( z z z -direction), causing crack faces to twist and tear apart.
Mode I Stress Intensity Factor ( K I K_I K I )
G.R. Irwin demonstrated that the asymptotic elastic stress field at distance r r r and angle θ \theta θ from a Mode I crack tip is governed by a single scalar parameter:
σ i j ( r , θ ) = K I 2 π r f i j ( θ ) + O ( r 0 ) \sigma_{ij}(r, \theta) = \frac{K_I}{\sqrt{2\pi r}} f_{ij}(\theta) + \mathcal{O}(r^0) σ ij ( r , θ ) = 2 π r K I f ij ( θ ) + O ( r 0 )
The Mode I Stress Intensity Factor ( K I K_I K I ) scales the entire local crack-tip stress field and is defined macroscopically as:
K I = Y σ π a K_I = Y \sigma \sqrt{\pi a} K I = Yσ πa
where:
σ \sigma σ is the nominal gross far-field tensile stress acting normal to the crack plane.
a a a is the characteristic crack dimension ( a = depth of an edge crack a = \text{depth of an edge crack} a = depth of an edge crack ; 2 a = total length of an internal crack 2a = \text{total length of an internal crack} 2 a = total length of an internal crack , so a = half-crack length a = \text{half-crack length} a = half-crack length ).
Y Y Y (also denoted β \beta β ) is the dimensionless geometry correction factor , which accounts for crack geometry, component thickness, finite width ( a / W a/W a / W ), and boundary constraints.
Central through-crack of length 2 a 2a 2 a in an infinitely wide plate: Y = 1.0 Y = 1.0 Y = 1.0 .
Single edge crack of depth a a a in a semi-infinite plate: Y = 1.12 Y = 1.12 Y = 1.12 (the 12 % 12\% 12% increase reflects the loss of mechanical constraint at the free edge).
Units: MPa m \text{MPa}\sqrt{\text{m}} MPa m in SI; ksi in \text{ksi}\sqrt{\text{in}} ksi in in USCS. Conversion factor: 1 ksi in = 1.09884 MPa m 1\text{ ksi}\sqrt{\text{in}} = 1.09884\text{ MPa}\sqrt{\text{m}} 1 ksi in = 1.09884 MPa m
Plane Strain Fracture Toughness ( K I c K_{Ic} K I c ) & Fracture Criterion
When the applied stress intensity factor reaches a critical threshold, crack growth becomes unstable and propagates through the component at high speed (a sizable fraction of the material's sound speed, on the order of 1 km/s 1\text{ km/s} 1 km/s in steel). This critical threshold under maximum triaxial constraint is the Plane Strain Fracture Toughness ( K I c K_{Ic} K I c ) —a fundamental material property measured experimentally in accordance with ASTM E399 .
The fracture mechanics design criterion is:
K I ≥ K I c ⟹ Unstable Brittle Fracture K_I \ge K_{Ic} \implies \text{Unstable Brittle Fracture} K I ≥ K I c ⟹ Unstable Brittle Fracture
The factor of safety against brittle fracture is:
n = K I c K I = K I c Y σ π a n = \frac{K_{Ic}}{K_I} = \frac{K_{Ic}}{Y \sigma \sqrt{\pi a}} n = K I K I c = Yσ πa K I c
ASTM E399 Valid Plane Strain Thickness Requirement
Fracture toughness is strongly dependent on specimen thickness ( B B B ):
Thin Plates (Plane Stress): At a free surface, out-of-plane normal stress is zero ( σ z = 0 \sigma_z = 0 σ z = 0 ). The material is free to contract laterally ( ϵ z ≠ 0 \epsilon_z \ne 0 ϵ z = 0 ), allowing extensive plastic yielding in a large yield zone ahead of the crack tip. This blunts the crack, dissipates substantial plastic work, and produces high apparent toughness ( K c K_c K c ).
Thick Plates (Plane Strain): Deep inside a thick plate, surrounding elastic material prevents lateral contraction ( ϵ z = 0 \epsilon_z = 0 ϵ z = 0 ). This constraint generates a large triaxial tensile stress ( σ z ≈ ν ( σ x + σ y ) \sigma_z \approx \nu(\sigma_x + \sigma_y) σ z ≈ ν ( σ x + σ y ) ). Triaxial tension severely suppresses plastic slip, producing a microscopic plastic zone and an invariant, lower-bound material toughness plateau: K I c K_{Ic} K I c .
To ensure that experimental measurements or design calculations are valid lower-bound plane strain values, textbooks use the classic ASTM E399 size criterion:
B ≥ 2.5 ( K I c S y ) 2 and a ≥ 2.5 ( K I c S y ) 2 B \ge 2.5\left(\frac{K_{Ic}}{S_y}\right)^2 \quad \text{and} \quad a \ge 2.5\left(\frac{K_{Ic}}{S_y}\right)^2 B ≥ 2.5 ( S y K I c ) 2 and a ≥ 2.5 ( S y K I c ) 2
where S y S_y S y is the 0.2 % 0.2\% 0.2% offset yield strength of the material. Since the 2012 revision, ASTM E399 makes the uncracked ligament the mandatory size check, W − a ≥ 2.5 ( K I c / σ Y S ) 2 W - a \ge 2.5(K_{Ic}/\sigma_{YS})^2 W − a ≥ 2.5 ( K I c / σ Y S ) 2 , together with standard specimen proportions (such as B ≈ W / 2 B \approx W/2 B ≈ W /2 ) and a P max / P Q ≤ 1.10 P_{\max}/P_Q \le 1.10 P m a x / P Q ≤ 1.10 limit. These rules qualify a laboratory test specimen ; in design, K I c K_{Ic} K I c is used as a conservative lower-bound property for thick, highly constrained parts.
Irwin Plastic Zone Size
Because actual engineering metals yield at stresses exceeding S y S_y S y , a small zone of plastic deformation develops at the crack tip. Irwin estimated the radius of this plastic zone ( r y r_y r y ) as:
r y = 1 2 π ( K I S y ) 2 ( Plane Stress ) r_y = \frac{1}{2\pi}\left(\frac{K_I}{S_y}\right)^2 \quad (\text{Plane Stress}) r y = 2 π 1 ( S y K I ) 2 ( Plane Stress )
r y = 1 6 π ( K I S y ) 2 ( Plane Strain ) r_y = \frac{1}{6\pi}\left(\frac{K_I}{S_y}\right)^2 \quad (\text{Plane Strain}) r y = 6 π 1 ( S y K I ) 2 ( Plane Strain )
Under plane strain, triaxial constraint reduces the plastic zone volume by a factor of 3, promoting sudden cleavage.
Non-Destructive Evaluation (NDE) Crack Detection Thresholds
In modern damage-tolerant aerospace and pressure vessel design, structures are assumed to contain initial flaws ( a 0 a_0 a 0 ) equal to the largest flaw that could reliably escape detection during quality inspection (typically established at a 90 % 90\% 90% Probability of Detection with a 95 % 95\% 95% confidence limit, 90/95 POD):
Inspection Methodology Acronym Typical Surface Flaw Limit ( a detect a_{\text{detect}} a detect ) Subsurface Flaw Limit ( a detect a_{\text{detect}} a detect )
Visual Testing VT 5.0 – 10.0 mm 5.0\text{--}10.0\text{ mm} 5.0 – 10.0 mm ( 0.20 – 0.40 in 0.20\text{--}0.40\text{ in} 0.20 – 0.40 in ) Not Detectable
Radiographic Testing RT 1.5 – 3.0 mm 1.5\text{--}3.0\text{ mm} 1.5 – 3.0 mm ( 0.06 – 0.12 in 0.06\text{--}0.12\text{ in} 0.06 – 0.12 in ) 1.5 – 3.0 mm 1.5\text{--}3.0\text{ mm} 1.5 – 3.0 mm ( 0.06 – 0.12 in 0.06\text{--}0.12\text{ in} 0.06 – 0.12 in )
Ultrasonic Testing UT 1.0 – 2.0 mm 1.0\text{--}2.0\text{ mm} 1.0 – 2.0 mm ( 0.04 – 0.08 in 0.04\text{--}0.08\text{ in} 0.04 – 0.08 in ) 1.0 – 2.0 mm 1.0\text{--}2.0\text{ mm} 1.0 – 2.0 mm ( 0.04 – 0.08 in 0.04\text{--}0.08\text{ in} 0.04 – 0.08 in )
Magnetic Particle Testing MT 0.5 – 1.5 mm 0.5\text{--}1.5\text{ mm} 0.5 – 1.5 mm ( 0.02 – 0.06 in 0.02\text{--}0.06\text{ in} 0.02 – 0.06 in ) Shallow only ( < 2 mm < 2\text{ mm} < 2 mm )
Liquid Penetrant Testing PT 0.5 – 1.5 mm 0.5\text{--}1.5\text{ mm} 0.5 – 1.5 mm ( 0.02 – 0.06 in 0.02\text{--}0.06\text{ in} 0.02 – 0.06 in ) Not Detectable
Eddy Current Testing ET 0.2 – 0.8 mm 0.2\text{--}0.8\text{ mm} 0.2 – 0.8 mm ( 0.008 – 0.03 in 0.008\text{--}0.03\text{ in} 0.008 – 0.03 in ) Near-surface only
Leak-Before-Break (LBB) Design Criterion
In the design of high-pressure piping, gas storage cylinders, and reactor pressure vessels (ASME Section XI), Leak-Before-Break (LBB) is the primary safety criterion against catastrophic explosion.
At a design operating stress σ \sigma σ , the critical crack size for unstable fast fracture is:
K I = K I c ⟹ Y σ π a c = K I c ⟹ a c = 1 π ( K I c Y σ ) 2 K_I = K_{Ic} \implies Y \sigma \sqrt{\pi a_c} = K_{Ic} \implies a_c = \frac{1}{\pi}\left(\frac{K_{Ic}}{Y \sigma}\right)^2 K I = K I c ⟹ Yσ π a c = K I c ⟹ a c = π 1 ( Yσ K I c ) 2
Let t t t represent the wall thickness of the pressure vessel:
Leak-Before-Break Satisfied ( a c > t a_c > t a c > t ): If the critical crack length exceeds the wall thickness, a growing crack will breach the exterior surface ( a = t a = t a = t ) while remaining stable. The pressurized fluid escapes through the through-wall slit, producing a detectable drop in system pressure or visible fluid venting. Operators can execute an orderly emergency shutdown before catastrophic fracture occurs.
Leak-Before-Break Violated ( a c < t a_c < t a c < t ): If the critical crack depth is less than the wall thickness, the crack reaches unstable fast-fracture velocity while still partially embedded inside the wall. The vessel ruptures catastrophically into multiple flying fragments without prior leakage warning.
- 5. Worked Numerical Engineering Scenario: Linear Elastic Fracture Mechanics of a Heavy Pressure Wall
- Problem Statement
A thick-walled tension plate forming part of a high-pressure hydraulic accumulator is fabricated from high-strength quenched and tempered alloy steel (AISI 4340). Quality control records document the following mechanical properties:
Yield strength: S y = 1200 MPa S_y = 1200\text{ MPa} S y = 1200 MPa ( 174 ksi 174\text{ ksi} 174 ksi )
Ultimate tensile strength: S u t = 1400 MPa S_{ut} = 1400\text{ MPa} S u t = 1400 MPa ( 203 ksi 203\text{ ksi} 203 ksi )
Plane strain fracture toughness: K I c = 65.0 MPa m K_{Ic} = 65.0\text{ MPa}\sqrt{\text{m}} K I c = 65.0 MPa m ( 59.2 ksi in 59.2\text{ ksi}\sqrt{\text{in}} 59.2 ksi in )
The plate has a rectangular cross-section with width W = 200 mm W = 200\text{ mm} W = 200 mm ( 0.200 m 0.200\text{ m} 0.200 m ) and thickness B = t = 25.0 mm B = t = 25.0\text{ mm} B = t = 25.0 mm ( 0.025 m 0.025\text{ m} 0.025 m ). It is subjected to a uniform nominal gross tensile stress of σ = 350 MPa \sigma = 350\text{ MPa} σ = 350 MPa ( 50.8 ksi 50.8\text{ ksi} 50.8 ksi ).
During a scheduled ultrasonic maintenance inspection (UT), an edge crack of depth a = 6.0 mm a = 6.0\text{ mm} a = 6.0 mm ( 0.0060 m 0.0060\text{ m} 0.0060 m ) is detected normal to the tensile stress axis. For this single-edge notched geometry, the geometry correction factor is governed by:
Y = 1.12 − 0.231 ( a W ) + 10.55 ( a W ) 2 − 21.72 ( a W ) 3 + 30.39 ( a W ) 4 Y = 1.12 - 0.231\left(\frac{a}{W}\right) + 10.55\left(\frac{a}{W}\right)^2 - 21.72\left(\frac{a}{W}\right)^3 + 30.39\left(\frac{a}{W}\right)^4 Y = 1.12 − 0.231 ( W a ) + 10.55 ( W a ) 2 − 21.72 ( W a ) 3 + 30.39 ( W a ) 4
Perform the following engineering evaluations:
Verify whether the component satisfies the ASTM E399 minimum thickness requirement for valid plane strain conditions.
Calculate the operating Mode I stress intensity factor ( K I K_I K I ) and the resulting factor of safety against catastrophic brittle fracture.
Determine the critical crack depth ( a c a_c a c ) that would cause sudden brittle fracture under the current operating stress.
Determine whether this component satisfies the Leak-Before-Break (LBB) safety criterion.
- Step-by-Step Solution
- Step 1: ASTM E399 Valid Plane Strain Thickness Verification
The minimum thickness required for valid plane strain fracture toughness behavior is:
B min = 2.5 ( K I c S y ) 2 B_{\min} = 2.5\left(\frac{K_{Ic}}{S_y}\right)^2 B m i n = 2.5 ( S y K I c ) 2
Substitute the material properties:
B min = 2.5 ( 65.0 MPa m 1200 MPa ) 2 = 2.5 ( 0.054167 m ) 2 = 2.5 ( 0.002934 m ) = 0.007335 m = 7.34 mm B_{\min} = 2.5\left(\frac{65.0\text{ MPa}\sqrt{\text{m}}}{1200\text{ MPa}}\right)^2 = 2.5(0.054167\sqrt{\text{m}})^2 = 2.5(0.002934\text{ m}) = 0.007335\text{ m} = 7.34\text{ mm} B m i n = 2.5 ( 1200 MPa 65.0 MPa m ) 2 = 2.5 ( 0.054167 m ) 2 = 2.5 ( 0.002934 m ) = 0.007335 m = 7.34 mm
Comparing with the actual plate dimensions:
Actual thickness: B = 25.0 mm B = 25.0\text{ mm} B = 25.0 mm , more than three times 7.34 mm 7.34\text{ mm} 7.34 mm , so the plate is thick enough for high crack-tip constraint.
Actual crack depth: a = 6.0 mm a = 6.0\text{ mm} a = 6.0 mm , slightly less than 7.34 mm 7.34\text{ mm} 7.34 mm , so a specimen this shallow would not meet the classic size rule. A shallow crack loses some constraint, which raises the apparent toughness, so using the lower-bound K I c K_{Ic} K I c is conservative.
The size criteria qualify a laboratory test; for this component the conclusion is that K I c = 65.0 MPa m K_{Ic} = 65.0\text{ MPa}\sqrt{\text{m}} K I c = 65.0 MPa m is an appropriate (conservative) design toughness.
Step 2: Operating Stress Intensity Factor ( K I K_I K I ) & Factor of Safety
First, determine the crack-to-width ratio:
a W = 6.0 mm 200.0 mm = 0.030 \frac{a}{W} = \frac{6.0\text{ mm}}{200.0\text{ mm}} = 0.030 W a = 200.0 mm 6.0 mm = 0.030
Calculate the geometry correction factor Y Y Y :
Y = 1.12 − 0.231 ( 0.030 ) + 10.55 ( 0.030 ) 2 − 21.72 ( 0.030 ) 3 + 30.39 ( 0.030 ) 4 Y = 1.12 - 0.231(0.030) + 10.55(0.030)^2 - 21.72(0.030)^3 + 30.39(0.030)^4 Y = 1.12 − 0.231 ( 0.030 ) + 10.55 ( 0.030 ) 2 − 21.72 ( 0.030 ) 3 + 30.39 ( 0.030 ) 4
Y = 1.12 − 0.00693 + 0.009495 − 0.000586 + 0.000025 = 1.1220 ≈ 1.122 Y = 1.12 - 0.00693 + 0.009495 - 0.000586 + 0.000025 = 1.1220 \approx 1.122 Y = 1.12 − 0.00693 + 0.009495 − 0.000586 + 0.000025 = 1.1220 ≈ 1.122
Now calculate the applied Mode I stress intensity factor:
K I = Y σ π a = ( 1.122 ) ( 350 MPa ) π ( 0.0060 m ) K_I = Y \sigma \sqrt{\pi a} = (1.122)(350\text{ MPa})\sqrt{\pi (0.0060\text{ m})} K I = Yσ πa = ( 1.122 ) ( 350 MPa ) π ( 0.0060 m )
π ( 0.0060 ) = 0.0188496 = 0.137294 m \sqrt{\pi (0.0060)} = \sqrt{0.0188496} = 0.137294\sqrt{\text{m}} π ( 0.0060 ) = 0.0188496 = 0.137294 m
K I = ( 1.122 ) ( 350 ) ( 0.137294 ) = 53.915 MPa m ≈ 53.92 MPa m K_I = (1.122)(350)(0.137294) = 53.915\text{ MPa}\sqrt{\text{m}} \approx 53.92\text{ MPa}\sqrt{\text{m}} K I = ( 1.122 ) ( 350 ) ( 0.137294 ) = 53.915 MPa m ≈ 53.92 MPa m
The factor of safety against static brittle fracture is:
n = K I c K I = 65.0 MPa m 53.92 MPa m = 1.205 ≈ 1.21 n = \frac{K_{Ic}}{K_I} = \frac{65.0\text{ MPa}\sqrt{\text{m}}}{53.92\text{ MPa}\sqrt{\text{m}}} = 1.205 \approx 1.21 n = K I K I c = 53.92 MPa m 65.0 MPa m = 1.205 ≈ 1.21
Note on Yielding: The nominal stress is σ = 350 MPa \sigma = 350\text{ MPa} σ = 350 MPa , giving a static tensile yield factor of safety n y = S y / σ = 1200 / 350 = 3.43 n_y = S_y / \sigma = 1200 / 350 = 3.43 n y = S y / σ = 1200/350 = 3.43 . However, the presence of the 6 mm 6\text{ mm} 6 mm crack reduces the true safety margin to a precarious 1.21 1.21 1.21 !
Step 3: Critical Crack Depth ( a c a_c a c ) Under Operating Stress
At the point of catastrophic fracture, K I = K I c = 65.0 MPa m K_I = K_{Ic} = 65.0\text{ MPa}\sqrt{\text{m}} K I = K I c = 65.0 MPa m . Using Y ≈ 1.12 Y \approx 1.12 Y ≈ 1.12 for shallow edge cracks:
K I c = Y σ π a c ⟹ a c = 1 π ( K I c Y σ ) 2 K_{Ic} = Y \sigma \sqrt{\pi a_c} \implies a_c = \frac{1}{\pi}\left(\frac{K_{Ic}}{Y \sigma}\right)^2 K I c = Yσ π a c ⟹ a c = π 1 ( Yσ K I c ) 2
a c = 1 π ( 65.0 MPa m ( 1.12 ) ( 350 MPa ) ) 2 = 1 π ( 65.0 392.0 ) 2 = 1 π ( 0.165816 ) 2 = 0.027495 π = 0.008752 m = 8.75 mm a_c = \frac{1}{\pi}\left(\frac{65.0\text{ MPa}\sqrt{\text{m}}}{(1.12)(350\text{ MPa})}\right)^2 = \frac{1}{\pi}\left(\frac{65.0}{392.0}\right)^2 = \frac{1}{\pi}(0.165816)^2 = \frac{0.027495}{\pi} = 0.008752\text{ m} = 8.75\text{ mm} a c = π 1 ( ( 1.12 ) ( 350 MPa ) 65.0 MPa m ) 2 = π 1 ( 392.0 65.0 ) 2 = π 1 ( 0.165816 ) 2 = π 0.027495 = 0.008752 m = 8.75 mm
Refining with a / W = 8.75 / 200 = 0.0438 a/W = 8.75/200 = 0.0438 a / W = 8.75/200 = 0.0438 and all five terms of the geometry polynomial gives Y = 1.128 Y = 1.128 Y = 1.128 :
a c = 1 π ( 65.0 ( 1.128 ) ( 350 ) ) 2 = 1 π ( 65.0 394.8 ) 2 = 1 π ( 0.16464 ) 2 = 0.00863 m = 8.63 mm a_c = \frac{1}{\pi}\left(\frac{65.0}{(1.128)(350)}\right)^2 = \frac{1}{\pi}\left(\frac{65.0}{394.8}\right)^2 = \frac{1}{\pi}(0.16464)^2 = 0.00863\text{ m} = 8.63\text{ mm} a c = π 1 ( ( 1.128 ) ( 350 ) 65.0 ) 2 = π 1 ( 394.8 65.0 ) 2 = π 1 ( 0.16464 ) 2 = 0.00863 m = 8.63 mm
(A second iteration changes nothing at this precision.) Catastrophic fracture will occur if the crack grows by an additional 8.63 − 6.00 = 2.63 mm 8.63 - 6.00 = 2.63\text{ mm} 8.63 − 6.00 = 2.63 mm .
Step 4: Leak-Before-Break (LBB) Evaluation
Treat the 25.0 mm 25.0\text{ mm} 25.0 mm plate thickness as the pressure wall and apply the simplified textbook LBB check, comparing the critical crack size with the wall thickness t t t (a full LBB assessment uses surface-crack and through-wall-crack solutions for the real vessel geometry):
a c = 8.63 mm < t = 25.0 mm a_c = 8.63\text{ mm} < t = 25.0\text{ mm} a c = 8.63 mm < t = 25.0 mm
Engineering Finding: The critical crack size ( 8.63 mm 8.63\text{ mm} 8.63 mm ) is far smaller than the 25 mm 25\text{ mm} 25 mm wall thickness ( a c / t = 0.345 a_c / t = 0.345 a c / t = 0.345 , roughly one-third of the wall).
Safety Conclusion: The component fails the Leak-Before-Break criterion. A growing crack would reach the fast-fracture size long before it could breach the wall and leak, so the wall would rupture without a warning leak.
Corrective Engineering Action: For the crack to reach the full 25 mm 25\text{ mm} 25 mm wall before going unstable ( Y ≈ 1.22 Y \approx 1.22 Y ≈ 1.22 at a / W = 0.125 a/W = 0.125 a / W = 0.125 ), the material would need K I c ≥ ( 1.22 ) ( 350 ) π ( 0.025 ) ≈ 120 MPa m K_{Ic} \ge (1.22)(350)\sqrt{\pi(0.025)} \approx 120\text{ MPa}\sqrt{\text{m}} K I c ≥ ( 1.22 ) ( 350 ) π ( 0.025 ) ≈ 120 MPa m at the present stress (for example, a tougher grade or a lower-strength temper of the steel). Alternatively, keeping K I c = 65 MPa m K_{Ic} = 65\text{ MPa}\sqrt{\text{m}} K I c = 65 MPa m , the operating stress would have to drop to about 65 / ( 1.22 π ( 0.025 ) ) ≈ 190 MPa 65/(1.22\sqrt{\pi(0.025)}) \approx 190\text{ MPa} 65/ ( 1.22 π ( 0.025 ) ) ≈ 190 MPa .