Materials Selection and Failure Theories

Introduction

Mechanical-system design is fundamentally a materials decision. A component may have adequate strength in a laboratory test yet fail in service because of fatigue, corrosion, wear, creep, stress concentration, poor processing, or an unsuitable manufacturing route. Materials selection therefore cannot be reduced to choosing the material with the highest strength or lowest price. It is a systems decision linking functional requirements, material properties, geometry, processing, environment, inspection, maintenance, sustainability, and cost.

A sound design process asks four connected questions:

1. What must the component do?
2. Which properties control performance and failure?
3. How will processing and service environment alter those properties?
4. Which material and manufacturing route provide acceptable performance at acceptable risk and cost?

This chapter develops the principal failure theories used in mechanical design and shows how material-selection charts and performance indices support decisions involving metals, polymers, ceramics, and composites.

Learning Objectives

By the end of this chapter, the reader should be able to:

• distinguish stiffness, strength, toughness, hardness, ductility, fatigue resistance, creep resistance, and wear resistance;
• relate material structure, processing, and environment to design performance;
• apply yielding, fracture, fatigue, creep, wear, and buckling criteria;
• use factor of safety and service-life concepts appropriately;
• interpret material-selection charts and derive simple performance indices;
• compare metals, polymers, ceramics, and composites for mechanical systems;
• justify a material choice using technical, manufacturing, economic, environmental, and safety criteria.

1. Material Properties and Design Requirements

1.1 Stiffness and strength

Stiffness describes resistance to elastic deformation. In uniaxial tension, Young’s modulus is

where E is Young’s modulus in pascals, σ is normal stress in pascals, and ε is dimensionless strain. This linear relation applies within the elastic range for a material behaving approximately elastically.

Strength describes resistance to permanent deformation or fracture. Important values include yield strength σᵧ, ultimate tensile strength σᵤ, compressive strength, and shear strength. Strength depends on temperature, loading rate, specimen geometry, defects, processing history, and environment.

Specific stiffness and specific strength are useful when mass matters:

Here ρ is density. These indices are meaningful only when the governing design constraint and geometry are also considered.

1.2 Ductility, toughness, and hardness

Ductility is the capacity for plastic deformation before fracture. It is valuable because plastic deformation can redistribute stress and provide visible warning. Toughness is the energy absorbed before fracture and is especially important under impact, low temperature, and crack-like defects. Fracture toughness, commonly denoted K_IC for plane-strain Mode I loading, measures resistance to unstable crack growth.

Hardness is resistance to indentation, scratching, or localized plastic deformation. It often correlates with wear resistance and strength, but it is not a substitute for toughness or fatigue data. Very hard materials may be brittle.

1.3 Functional and environmental requirements

A design specification should identify:

• loads: static, cyclic, impact, vibration, pressure, and thermal;
• allowable deformation and dimensional tolerance;
• service temperature and thermal cycling;
• moisture, chemicals, salt, radiation, and vacuum;
• required life and reliability;
• manufacturing process and production quantity;
• joining, repair, inspection, and recycling requirements.

Properties are not independent. Increasing strength by heat treatment may reduce ductility; increasing fiber content may increase stiffness while reducing impact tolerance; plasticizers may improve flexibility while reducing strength and increasing environmental sensitivity.

2. Structure, Processing, and Environment

Material performance is controlled by structure at several scales. In metals, grain size, phase distribution, dislocations, inclusions, and residual stress are important. In polymers, molecular weight, crystallinity, chain orientation, additives, and cross-linking dominate behavior. In ceramics, porosity, grain size, impurities, and surface flaws are critical. In composites, fiber orientation, volume fraction, interface quality, void content, and laminate architecture govern properties.

Processing can create beneficial or harmful features. Rolling may align grains and improve directional strength. Heat treatment can produce desired phases but may generate distortion or residual stress. Casting can be economical but may introduce shrinkage cavities and inclusions. Additive manufacturing can produce complex geometry but may create anisotropy, lack-of-fusion defects, and surface roughness.

The service environment can change a material continuously. Corrosion reduces cross-section and creates pits that act as fatigue initiators. Elevated temperature promotes creep and oxidation. Moisture can plasticize polymers and weaken composite interfaces. Hydrogen can cause embrittlement in susceptible alloys. Low temperature can reduce ductility in some materials and increase brittleness in polymers.

Consequently, handbook properties are starting points, not universal constants. Design values should be obtained for the relevant condition, including heat treatment, orientation, temperature, frequency, surface finish, and environment.

3. Static Failure and Yielding Theories

3.1 Factor of safety

For a simple tensile design, the factor of safety may be defined as

where n is the factor of safety, σ_allowable is the allowable stress, and σ_working is the calculated service stress. If yield governs, σ_allowable may be σᵧ divided by a prescribed design factor. The selected value depends on uncertainty, consequence of failure, variability, codes, inspection, and load knowledge.

3.2 Maximum normal stress theory

The maximum normal stress theory predicts yielding when the largest principal stress reaches the uniaxial yield strength. It is mainly suitable for brittle materials when tensile and compressive strengths are considered separately. It is generally conservative or inappropriate for ductile metals under multiaxial stress.

3.3 Tresca theory

The maximum shear stress theory states that yielding begins when the maximum shear stress reaches the shear stress at yield in a uniaxial test:

where σ₁ and σ₃ are the largest and smallest principal stresses. Equivalently, the Tresca equivalent stress is

Yielding is predicted when σ_T ≥ σᵧ. Tresca is appropriate for isotropic, ductile materials under static loading and is relatively conservative.

3.4 von Mises theory

The distortion-energy theory predicts yielding when the distortion energy reaches that at uniaxial yield. The equivalent stress is

For plane stress with σ₃ = 0,

Yielding is predicted when σ_VM ≥ σᵧ. The theory is widely used for ductile metals because it agrees well with yielding experiments under proportional loading. Both Tresca and von Mises assume homogeneous, approximately isotropic material behavior and do not predict fracture of brittle solids.

Worked Example 1: Combined stress

A steel point in a component has σ_x = 120 MPa, σ_y = 40 MPa, and τ_xy = 30 MPa. The yield strength is 300 MPa. For plane stress,

σ_VM = √(120² − 120×40 + 40² + 3×30²)

= √(14,400 − 4,800 + 1,600 + 2,700)

= √13,900 = 117.9 MPa.

The yield factor based on von Mises stress is

n = 300/117.9 = 2.54.

Under the stated static, elastic assumptions, yielding is not predicted. This result does not establish safety against fatigue, fracture, corrosion, or buckling.

4. Fracture and Crack Growth

A crack magnifies local stress. Linear-elastic fracture mechanics uses the stress-intensity factor

where K_I is Mode I stress intensity, Y is a dimensionless geometry factor, σ is nominal tensile stress, and a is crack size. Unstable fracture is predicted when

where K_IC is plane-strain fracture toughness. The equation is applicable when small-scale yielding and linear elasticity are reasonable. For large plastic zones, elastic-plastic parameters such as the J-integral or crack-tip opening displacement are more appropriate.

Fracture-critical design requires control of defects, suitable inspection intervals, damage-tolerant analysis, and consideration of residual stress. Ceramics and hardened steels often have high strength but low tolerance of flaws. Ductile alloys may tolerate cracks better but can still fail rapidly under high stress or aggressive environments.

5. Fatigue Failure

Fatigue is progressive damage under repeated or fluctuating stress. It may occur below the static yield strength. Important variables include stress amplitude σ_a, mean stress σ_m, stress ratio R, surface finish, size, temperature, residual stress, notches, corrosion, and frequency.

For constant-amplitude loading,

A commonly used mean-stress correction is the Goodman relation:

where Se is corrected endurance strength, σ_u is ultimate tensile strength, and n is the desired fatigue factor of safety. The linear Goodman relation is an engineering approximation for tensile mean stress; alternatives include Gerber and Soderberg relations. Soderberg uses yield strength in place of ultimate strength and is more conservative.

For finite-life fatigue, stress-life S–N curves are used for high-cycle conditions. Strain-life methods are more suitable when cyclic plasticity occurs. Crack-growth methods use relationships such as the Paris law:

where a is crack length, N is number of cycles, C and m are experimentally determined constants, and ∆K is the stress-intensity range. The Paris law applies over the stable mid-range of crack growth, not near the threshold or final rapid-fracture region.

Good fatigue design reduces stress concentration, improves surface finish, introduces beneficial compressive residual stress where appropriate, avoids sharp keyways, and validates the actual load spectrum.

6. Creep and High-Temperature Failure

Creep is time-dependent deformation under sustained stress, usually significant above approximately 0.3–0.4 of the absolute melting temperature for metals, though the threshold depends strongly on material and service duration. Creep has primary, secondary, and tertiary stages. The secondary stage is often approximated by a steady creep rate.

A useful empirical relationship is Norton’s law:

where ε̇_c is steady creep strain rate, A and n are temperature- and material-dependent constants, and σ is stress. Temperature dependence is often represented by an Arrhenius factor, but constants must be obtained from appropriate data.

Design against creep may involve limiting stress, limiting strain, specifying rupture life, using creep-resistant alloys, reducing temperature, or providing cooling. Polymers can creep substantially at ordinary temperatures, especially near their glass-transition temperature. Ceramics generally resist creep better at moderate temperatures but may suffer oxidation, thermal shock, or brittle fracture.

7. Wear and Surface Damage

Wear is material loss or surface damage caused by contact and relative motion. Common mechanisms include adhesive wear, abrasive wear, erosive wear, fretting, fatigue wear, and tribochemical wear.

The Archard wear relation is

where V is wear volume, k is a dimensionless wear coefficient, W is normal load, L is sliding distance, and H is hardness of the softer material. The relation is an approximate model for adhesive or abrasive wear under specified contact conditions; k is not universal and may change by orders of magnitude with lubrication, surface finish, temperature, and material pair.

Wear control can involve increasing hardness, selecting compatible material pairs, improving lubrication, filtering contaminants, reducing contact pressure, using coatings, and preventing misalignment. A hard material is not automatically a good tribological material: friction, thermal conductivity, toughness, transfer-film formation, and lubrication compatibility also matter.

8. Buckling of Slender Members

A column may fail by instability before its material yield strength is reached. For an ideal, straight, pin-ended elastic column, Euler’s critical load is

where E is Young’s modulus, I is the least second moment of area, L is unsupported length, and K is the effective-length factor. The critical stress is

where r = √(I/A) is the radius of gyration and A is cross-sectional area.

Euler’s equation assumes a slender, initially straight member, elastic material, ideal loading, and specified end conditions. Real columns have imperfections, residual stress, eccentricity, and inelastic behavior. Intermediate-slenderness columns require an inelastic formula or design-code curve. Increasing stiffness, reducing unsupported length, adding bracing, and increasing the least-axis moment of area improve buckling resistance.

9. Material Families and Selection

Material selection begins with a constraint. Examples include minimum stiffness, allowable stress, maximum mass, minimum fracture toughness, maximum operating temperature, or minimum wear life. A material index ranks candidates for a particular geometry and objective.

For a light, axially loaded tie of required length L and force F, with allowable stress σ_allow and density ρ, mass is

and the area required by strength is A = F/σ_allow. Therefore,

For minimum mass under strength control, maximize σ_allow/ρ. If stiffness controls elongation, δ = FL/(AE), so A = FL/(Eδ), and

Thus, maximize E/ρ. These indices apply only to the stated geometry, loading, constraint, and failure mode.

Material-selection charts

Ashby-style charts plot one property against another, often on logarithmic axes. Families of materials occupy characteristic regions. Selection proceeds by:

1. translating the design into objective, constraints, and free variables;
2. deriving the relevant performance index;
3. plotting or screening materials using the index;
4. removing materials incompatible with temperature, environment, processing, or safety;
5. comparing cost, availability, joining, inspection, and sustainability;
6. validating finalists through testing or qualified data.

Charts support screening; they do not replace engineering judgment. Property scatter, anisotropy, processing route, supply chain, and certification may dominate the final decision.

10. Processing and Manufacturing Compatibility

A material is suitable only if it can be manufactured into the required geometry with acceptable defects and cost. Casting favors complex metal shapes but requires control of solidification. Forging improves directional strength and reduces porosity but may require expensive tooling. Machining provides precision but wastes material and may damage surfaces. Injection molding is efficient for high-volume polymers but requires suitable flow and mold design. Composite lay-up permits directional properties but demands control of fiber placement, cure, voids, and bonding.

Design for manufacture should address tolerances, draft angles, radii, weld accessibility, heat-treatment distortion, residual stress, joining compatibility, and inspection access. A technically superior material may be rejected if it cannot be reliably produced or repaired.

11. Illustrative Case: Lightweight Drive Shaft

Background

A manufacturer is redesigning a rotating drive shaft for a mobile machine. The shaft transmits torque, experiences bending from gear forces, must operate in a wet and dusty environment, and should reduce mass without sacrificing fatigue life.

Problem

A steel shaft meets static strength requirements but is heavy. Candidate alternatives are aluminum alloy, glass-fiber-reinforced polymer, and carbon-fiber/epoxy composite. The shaft also requires reliable torque transfer at metallic couplings.

Analysis

Static strength alone is insufficient. The design must consider torsional stiffness, fatigue under variable torque, environmental moisture, bearing-seat wear, galvanic corrosion, impact damage, joining, and inspection. A carbon-fiber composite offers high specific stiffness and strength, but its properties depend strongly on fiber direction and laminate design. A simple isotropic chart comparison could therefore be misleading.

The engineering team defines the shaft’s torsional stiffness, allowable angular twist, fatigue spectrum, maximum speed, temperature, and target life. The composite tube is designed with fibers oriented at ±45° for torsional load, while hoop and axial plies support other loads and protect against damage. Metallic end fittings are bonded and mechanically secured. The design includes electrical isolation to reduce galvanic corrosion and protective coating against moisture and abrasion.

Approach and validation

Finite-element analysis estimates stresses, deformation, modal frequencies, and interlaminar stresses. Coupon tests characterize the laminate, adhesive, and environmental conditioning. Full-scale rotating tests reproduce torque spectrum, imbalance, temperature, and moisture. Non-destructive inspection is specified for bonding and delamination.

Outcome

The composite design reduces mass, but the final selection depends on life-cycle cost and inspection capability. Aluminum may be preferred for moderate production volume and simpler repair. The composite is preferred when mass reduction and torsional stiffness justify higher manufacturing and inspection costs.

Learning points

The case demonstrates that material selection is a coupled decision involving failure mode, architecture, processing, joining, environment, and verification. It also shows why “strongest material” is not a meaningful selection criterion without a defined load case and design objective.

12. Design Practice, Safety, and Contemporary Developments

Engineering decisions should be traceable. A material-selection record should identify requirements, data sources, assumptions, safety factors, environmental corrections, rejected alternatives, tests, and applicable standards. Safety-critical components require qualified materials, controlled processes, inspection, configuration management, and documented change control.

Sustainability should include embodied energy, service life, repairability, recyclability, toxicity, and end-of-life pathways. Lightweighting can reduce operational energy, but difficult-to-recycle composites may create end-of-life challenges. Material substitution must not shift environmental burden without analysis.

Computational materials databases, machine learning, and generative design can accelerate screening. Their results remain dependent on data quality and physical constraints. AI-generated recommendations should be checked against certified property data, manufacturability, uncertainty, and failure-mechanism models. Explainability and traceability are particularly important in safety-critical design.

Best practice is to use a staged process: screen broadly, eliminate incompatible candidates, rank using a justified index, analyze governing failure modes, test under representative conditions, and monitor field performance. Selection should be revisited when loads, suppliers, processes, or environments change.

Chapter Summary

Materials selection links component function, mechanical properties, processing, environment, manufacturing, safety, cost, and sustainability. Stiffness controls deformation; strength controls yielding; toughness and fracture mechanics address crack tolerance; fatigue methods address cyclic loading; creep models address long-term high-temperature deformation; wear models address surface damage; and buckling theory addresses instability of slender members.

Tresca and von Mises criteria are primarily used for yielding in ductile materials, while brittle materials require attention to principal stress, fracture toughness, and flaw populations. Material-selection indices such as E/ρ and σ/ρ are derived from a specified objective and constraint and must not be used outside their assumptions. Metals, polymers, ceramics, and composites each offer distinct combinations of performance and limitation. A defensible choice is validated through processing analysis, environmental assessment, testing, inspection, and life-cycle consideration.

Key Terms

Factor of safety: Ratio of allowable capacity to working demand.

Young’s modulus: Elastic stiffness in uniaxial loading.

Yield strength: Stress marking the onset of significant plastic deformation.

Ductility: Capacity for plastic deformation before fracture.

Toughness: Energy absorption capacity before fracture.

Hardness: Resistance to indentation or localized deformation.

Fracture toughness: Resistance to crack growth under defined conditions.

Stress intensity factor: Crack-tip parameter describing the severity of a crack-loading field.

Fatigue: Progressive damage caused by repeated or fluctuating loading.

Stress amplitude: Half the difference between maximum and minimum cyclic stress.

Creep: Time-dependent deformation under sustained stress.

Wear: Material loss or surface damage from contact and motion.

Buckling: Instability involving sudden lateral deformation of a compressed member.

Specific strength: Strength divided by density.

Material index: Property combination used to rank materials for a defined design objective.

Anisotropy: Direction-dependent material behavior.

Residual stress: Stress remaining after manufacture or prior loading.

Delamination: Separation between layers of a laminated composite.

Design value: Property value adjusted for condition, uncertainty, and required reliability.

Review Questions

Short Answer Questions

1. Distinguish stiffness from strength and explain why a strong material may still produce excessive deformation.
2. State the assumptions and purpose of the von Mises yielding criterion.
3. Define stress amplitude, mean stress, and stress ratio for cyclic loading.
4. Why are cracks particularly important in ceramics and hardened metals?
5. Explain why material-selection charts cannot by themselves determine a final material.

Long Answer Questions

1. Compare Tresca and von Mises theories for ductile-metal design, including their equations, assumptions, and relative conservatism.
2. Explain how structure, processing, and service environment affect the performance of metals, polymers, ceramics, and composites.
3. Describe a complete fatigue-design procedure for a rotating shaft subject to variable bending and torsion.
4. Discuss creep, wear, and buckling as failure modes that may occur without conventional tensile yielding.
5. Develop a defensible material-selection procedure for a safety-critical mechanical component, including testing and documentation.

Analytical / Application Questions

1. A component experiences plane stress σ_x = 150 MPa, σ_y = −30 MPa, and τ_xy = 25 MPa. Calculate the von Mises stress and determine the factor of safety if the yield strength is 360 MPa.
2. A slender pin-ended column has E = 200 GPa, I = 2.0 × 10⁻⁸ m⁴, and L = 1.0 m. Calculate its Euler critical load and state whether the result is applicable without checking slenderness and imperfections.
3. A designer must select a material for a lightweight tension member whose performance is strength-controlled. Explain which material index should be maximized and compare the implications of selecting aluminum, a polymer composite, or steel.

Multiple Choice Questions with Answers

Question 1

Which property primarily measures resistance to elastic deformation?

A. Young’s modulus

B. Fracture toughness

C. Hardness

D. Ductility

Correct Answer: A. Young’s modulus

Question 2

For a ductile material under plane stress, which criterion is widely used to predict yielding?

A. Archard criterion

B. von Mises criterion

C. Paris criterion

D. Euler criterion

Correct Answer: B. von Mises criterion

Question 3

In the equation K_I = Yσ√(πa), what does a represent?

A. Material density

B. Crack size

C. Stress amplitude

D. Creep exponent

Correct Answer: B. Crack size

Question 4

The Goodman relation is primarily used in the design of components subjected to:

A. Cyclic stress with a mean stress

B. Pure thermal expansion

C. Static hydrostatic pressure only

D. Elastic column instability only

Correct Answer: A. Cyclic stress with a mean stress

Question 5

According to the ideal Euler equation, increasing the unsupported length of a column causes the critical load to:

A. Increase linearly

B. Remain unchanged

C. Decrease with the square of length

D. Increase with the square of length

Correct Answer: C. Decrease with the square of length

Question 6

In the Archard wear relation, increasing the hardness of the softer material generally causes predicted wear volume to:

A. Increase proportionally

B. Decrease, all other variables unchanged

C. Become independent of load

D. Equal the sliding distance

Correct Answer: B. Decrease, all other variables unchanged

Question 7

For minimum mass of a strength-controlled tension member, the most relevant simple material index is:

A. ρ/σ_allow

B. σ_allow/ρ

C. Eρ

D. H/E

Correct Answer: B. σ_allow/ρ

Question 8

Which material family is generally most sensitive to flaw-controlled brittle fracture?

A. Ductile low-carbon steel

B. Elastomeric polymer

C. Structural ceramic

D. Annealed copper

Correct Answer: C. Structural ceramic

Question 9

Norton’s law relates steady creep strain rate primarily to:

A. Stress through a material-dependent exponent

B. Crack length through geometry factor only

C. Hardness through sliding distance only

D. Density through column length

Correct Answer: A. Stress through a material-dependent exponent

Question 10

Why can a composite material not be selected reliably from a single isotropic strength value?

A. Composites have no measurable properties

B. Composite behavior depends on direction, laminate architecture, interfaces, and defects

C. Composites cannot experience fatigue

D. Composite density is always greater than steel

Correct Answer: B. Composite behavior depends on direction, laminate architecture, interfaces, and defects