Brittle Fracture and Impact Testing: Part One

---

The Brittle Fracture Problem: Historical Context

During World War II, the engineering community faced a critical challenge when numerous welded Liberty ships and T-2 tankers experienced catastrophic brittle failures. Some vessels broke completely in two, while others sustained partial damage. These failures predominantly occurred during winter months, affecting ships in both heavy seas and while anchored at dock. These incidents highlighted how normally ductile mild steel can become dangerously brittle under specific conditions.

These calamities prompted a comprehensive research program to identify causes and develop preventive measures. While research initially focused on addressing the immediate crisis, it expanded to enhance understanding of brittle fracture mechanisms more generally. Although the ship failures brought significant attention to brittle failure in mild steel, this problem extends well beyond maritime applications. Documented brittle failures in tanks, pressure vessels, pipelines, and bridges date back to 1886.

Fundamental Factors in Brittle Fracture

Three fundamental factors contribute to brittle-cleavage type fractures:

1.   A triaxial state of stress
2.   Low temperature
3.   High strain rate or rapid rate of loading

Importantly, all three factors need not exist simultaneously to produce brittle fracture. Most service failures of the brittle type result from a combination of triaxial stress state (such as exists at a notch) and low temperature. However, since these effects intensify at high loading rates, various impact tests have been developed to assess materials' susceptibility to brittle behavior. Steels with identical properties when tested in tension or torsion at slow strain rates can demonstrate significant differences in brittle fracture tendency when subjected to notched-impact testing.

Welding Considerations in Brittle Fracture

Since the ship failures occurred primarily in welded structures, welding was initially considered unsuitable for applications where brittle fracture might occur. Subsequent research has demonstrated that welding itself is not inherently inferior to other construction methods regarding brittle fracture resistance. However, strict quality control is essential to prevent weld defects that can act as stress raisers or notches. Modern electrodes now make it possible to create welds with superior properties compared to mild-steel plate.

Welded structures require more critical design considerations than equivalent riveted structures. Considerable effort has gone into developing safe designs for welded structures, emphasizing the elimination of stress raisers and avoiding excessive structural rigidity. During wartime shipbuilding, riveted sections known as "crack arresters" were incorporated to prevent brittle failures from propagating completely through the structure.

Notched-Bar Impact Testing Methods

Various notched-bar impact tests help determine a material's tendency toward brittle behavior. These tests detect material differences not observable in standard tension tests. Though the results cannot be easily expressed in design requirements due to the complexity of triaxial stress conditions at notches, they provide valuable comparative data.

While numerous notched-bar test specimen designs have been used to investigate brittle fracture, two classes have been standardized: Charpy specimens (predominant in the United States) and Izod specimens (favored in Great Britain).

The Charpy specimen features a square cross-section (10×10 mm) with a 45° V notch, 2 mm deep with a 0.25 mm root radius. The specimen is horizontally supported as a beam and loaded behind the notch by a heavy swinging pendulum. This forces the specimen to bend and fracture at a high strain rate (approximately 10³ s⁻¹). The Izod specimen, rarely used today, has either circular or square cross-section with a V notch near the clamped end.

The primary measurement from impact testing is the energy absorbed during specimen fracture. After breaking the test bar, the pendulum rebounds to a height inversely related to the energy absorbed. This energy, usually expressed in joules, is read directly from a calibrated dial on the impact tester.

Notched-bar impact tests provide the most meaningful data when conducted across a temperature range, enabling determination of the ductile-to-brittle transition temperature.

Advantages of the Charpy V-Notch Test

The Charpy V-notch impact test offers several advantages:

• Relatively simple test procedure
• Economical, small test specimens
• Easily conducted over a range of sub-ambient temperatures
• Well-suited for measuring notch toughness differences in low-strength materials like structural steels
• Valuable for comparing alloy composition and heat treatment effects on notch toughness
• Widely used for quality control and material acceptance purposes

The Instrumented Charpy Test

The standard Charpy test measures total energy absorbed during specimen fracture. Enhanced information can be obtained by instrumenting the impact tester to provide a load-line history during testing. This record enables determination of:

• Energy required for fracture initiation
• Energy required for fracture propagation
• Load for general yielding
• Maximum load
• Fracture load

Recent trends include using standard Charpy specimens pre-cracked by introducing a fatigue crack at the V-notch tip. These pre-cracked specimens, used in instrumented Charpy tests, facilitate measurement of dynamic fracture toughness values (K_Id).

Significance of Transition-Temperature Curves

The primary engineering application of Charpy testing is selecting materials resistant to brittle fracture through transition-temperature curves. This design philosophy aims to select materials with sufficient notch toughness under severe service conditions, allowing load-carrying capability calculations using standard strength of materials methods without considering fracture properties or stress concentration effects from cracks or flaws.

Transition-temperature behavior across materials falls into three categories:

1. Medium- and low-strength face-centered cubic (fcc) metals and most hexagonal close-packed (hcp) metals exhibit high notch toughness, making brittle fracture unlikely except in special reactive chemical environments.
2. High-strength materials (σ₀ > E/150) demonstrate low notch toughness, making brittle fracture possible at nominal stresses in the elastic range at all temperatures and strain rates when flaws are present. High-strength steel, aluminum, and titanium alloys belong to this category. At low temperatures, fracture occurs through brittle cleavage; at higher temperatures, through low-energy rupture. These conditions benefit from fracture mechanics analysis.
3. Low- and medium-strength body-centered cubic (bcc) metals, along with beryllium, zinc, and ceramic materials, show notch toughness strongly dependent on temperature. At low temperatures, fracture occurs through cleavage; at high temperatures, through ductile rupture. This creates a transition from notch brittle to notch tough behavior as temperature increases. In metals, this transition occurs at 0.1 to 0.2 of the absolute melting temperature (Tm), while in ceramics, it occurs at approximately 0.5 to 0.7 Tm.

A well-defined criterion bases the transition temperature on when fracture becomes 100 percent cleavage. This point, known as nil ductility temperature (NDT), represents the temperature at which fracture initiates with essentially no prior plastic deformation. Below the NDT, the probability of ductile fracture is negligible.

Metallurgical Factors Affecting Transition Temperature

Changes in chemical composition or microstructure can alter transition temperature in mild steel by over 55°C (100°F). Carbon and manganese content have the most significant impact, with transition temperature lowering approximately 5.5°C (10°F) for each 0.1 percent manganese increase. Carbon content also substantially affects maximum energy and energy transition-temperature curve shapes.

For satisfactory notch toughness, the manganese-to-carbon (Mn/C) ratio should be at least 3:1. A maximum decrease of about 55°C (100°F) in transition temperature is possible with higher Mn/C ratios.

Phosphorus strongly raises transition temperature. Nitrogen's role is complex due to its interaction with other elements but is generally considered detrimental to notch toughness. Nickel benefits notch toughness in amounts up to 2 percent, particularly in lowering the ductility transition temperature. Silicon, in amounts exceeding 0.25 percent, appears to raise the transition temperature. Molybdenum raises the transition almost as rapidly as carbon, while chromium has minimal effect.

Oxygen content critically influences notch toughness. In high-purity iron, oxygen contents above 0.003 percent produce intergranular fracture and corresponding low energy absorption.

Grain size strongly affects transition temperature. Decreasing grain diameter by one ASTM number (increasing the ASTM grain size number) reduces transition temperature by approximately 16°C (30°F) in mild steel. Decreasing grain diameter from ASTM grain size 5 to 10 can lower the 10 ft-lb Charpy V-notch transition temperature from about 39°C to -33°C (70°F to -60°F).

For alloy steels, energy absorbed during impact testing at a given temperature generally increases with increasing tempering temperature. However, a minimum typically occurs between 200-320°C (400-600°F). This phenomenon, previously called "260°C (500°F) embrittlement," is more appropriately termed "tempered-martensite embrittlement" since the temperature varies with steel composition and tempering time.

Drop-Weight Test and Other Large-Scale Tests

A primary limitation of the Charpy impact test is that its small specimen may not realistically model actual conditions. Small specimens not only produce considerable data scatter but also cannot provide the same constraint found in structures with much greater thickness.

To address this limitation, tests capable of handling specimens several inches thick have been developed, primarily by Pellini and colleagues at the Naval Research Laboratory. This need arose from difficulties in producing fracture in small laboratory specimens that would accurately predict performance in full-scale structures.

The first development was the explosion-crack-starter test featuring a short, brittle weld bead deposited on a 14×14×1-inch steel plate surface. The plate was placed over a circular die and dynamically loaded with an explosive charge. The brittle weld bead introduces a small natural crack similar to a weld-defect crack. Tests conducted over a temperature range determine various transition temperatures by examining fracture appearance.

Below the NDT, a flat fracture runs completely to the test plate edges. Above the nil ductility temperature, a plastic bulge forms in the plate center, though the fracture remains flat and elastic to the plate edge. At higher temperatures, the fracture does not propagate beyond the bulged region. The temperature at which elastic fracture no longer reaches the plate edge is called the fracture transition elastic (FTE), marking the highest temperature of purely elastic stress fracture propagation. At even higher temperatures, extensive plasticity creates a helmet-type bulge. The temperature above which this fully ductile tearing occurs is the fracture transition plastic (FTP).