The Mechanisms of Metallurgical Failure

On the Origin of Fracture

John Campbell

• Publisher: Butterworth-Heinemann
• Year: 2020
• Print ISBN: 9780128224113
• eBook ISBN: 9780128226797
• Topic: Technology & Engineering
• Subtopic: Materials Science
• Index: Technology & Engineering

Metallurgy of Fracture: The Mechanics of Metal Failurelooks at the origin of metal defects, their related mechanisms of failure, and the modification of casting procedures to eliminate these defects, clearly connecting the strength and durability of metals with their fabrication process. The book starts with a focus on the fracture of liquids, looking at topics such as homogeneous and heterogeneous nucleation, entrainment processes in bifilms and bubbles, furling and unfurling, ingot casting, continuous casting, and more. From there it discusses fracture of liquid and solid state, focusing on topics such as externally and internally initiated tearing. The book then concludes with a section discussing fracture of solid metals covering concepts such as ductility and brittleness, dislocation mechanisms, the relationship between the microstructure and properties of metals, corrosion, hydrogen embrittlement, and more. Improved approaches to fabrication and casting processes that will help eliminate these defects are provided throughout.- Looks at how the fracture of metals originates in the liquid-state due to poor casting practices- Offers improved casting techniques to reduce liquid-state borne fracture- Draws attention to the parallels between fracture initiation in the liquid and solid states- Covers spall tests and how to improve material quality by hot isostatic pressing

Chapter 1

The fracture of liquids

Abstract

The fundamental process of the consolidation of metals ensures bifilms are created and are widespread in engineering metals. The bifilm is a crack resulting from the impingement of two oxide films. The impingement of oxides on powder particles during the consolidation of powders to make powder metallurgical products is easily understood, but results in relatively harmless and controlled dispersions of cracks. Bifilms formed during the casting of liquid metals are quite different, ranging in sizes from submicrometer to fractions of meters. The many phenomena associated with their creation and their morphological changes with temperature and time are presented. The behaviors of the light metals and the higher temperature metals, particularly irons, steels, and Ni-based alloys, are contrasted. Techniques for the avoidance of bifilm formation are outlined. The macroscopic bifilm cracks, fundamental to current vacuum arc remelted steels, are a special danger, which are highlighted.

Keywords

Bifilms; Defects; Entrainment; ESR; Fracture; Oxide films; Turbulence; VAR

1. 1.1 Theoretical strength of liquids
   1. 1.1.1 Classical continuum theory
   2. 1.1.2 Classical bubble nucleation
   3. 1.1.3 Homogeneous nucleation
      1. 1.1.3.1 Heterogeneous nucleation
   4. 1.1.4 Nucleation conditions for shrinkage pores
   5. 1.1.5 Joint gas and shrinkage conditions
   6. 1.1.6 Pseudo-heterogeneous nucleation of pores
2. 1.2 Experimental demonstration of hydrostatic tensions in liquids
3. 1.3 Nonclassical pore formation mechanisms
   1. 1.3.1 High energy radiation
   2. 1.3.2 Preexisting suspension of bubbles
4. 1.4 Entrainment processes
   1. 1.4.1 Entrainment of bifilms
      1. 1.4.1.1 Visual evidence for bifilms
      2. 1.4.1.2 Surface turbulence
   2. 1.4.2 Weber Number We
   3. 1.4.3 Froude number Fr
   4. 1.4.4 Reynolds number Re
   5. 1.4.5 Oxide skins from melt charge materials
   6. 1.4.6 Pouring
   7. 1.4.7 The critical fall height
   8. 1.4.8 The oxide lap from surface flooding
   9. 1.4.9 Oxide lap as a confluence weld
   10. 1.4.10 The oxide flow tube
   11. 1.4.11 Microjetting
   12. 1.4.12 Entrainment of bubbles
       1. 1.4.12.1 Bubble trails
       2. 1.4.12.2 Bubble damage
   13. 1.4.13 Other entrainment defects
       1. 1.4.13.1 Extrinsic inclusions
       2. 1.4.13.2 Flux and slag inclusions
   14. 1.4.14 Furling and unfurling
       1. 1.4.14.1 Inflation by gas
       2. 1.4.14.2 Expansion by solidification shrinkage (3-D strain)
       3. 1.4.14.3 Transgranular straightening (flattening) by dendrite growth
       4. 1.4.14.4 Intergranular straightening by grains
       1.4.14.5 Straightening (flattening) by intermetallics and second phases >
       1. 1.4.14.6 Flattening by rigid flat intermetallic or second phase
       2. 1.4.14.7 Opening by 1-D cooling strain
       3. 1.4.14.8 Opening under service stress
   15. 1.4.15 Detrainment
       1. 1.4.15.1 Detrainment techniques take a variety of forms
   16. 1.4.16 Deactivation of entrained films
       1. 1.4.16.1 Loss of gas
       2. 1.4.16.2 Closing of bifilms by pressure
       3. 1.4.16.3 Bonding by diffusion reaction
       4. 1.4.16.4 Bonding by liquid binder
   17. 1.4.17 Morphological transformations
   18. 1.4.18 Soluble, transient films
   19. 1.4.19 Liquid Oxide Entrainment (inclusion shape control)
   20. 1.4.20 Nonprotective and unstable oxides
5. 1.5 Entrainment avoidance
6. 1.6 The quest for clean steels
   1. 1.6.1 Ingot casting
   2. 1.6.2 Continuous casting
      1. 1.6.2.1 Nozzle design
   3. 1.6.3 Secondary remelting processes
      1. 1.6.3.1 Vacuum induction melting (VIM)
      2. 1.6.3.2 Vacuum arc remelting (VAR)
      3. 1.6.3.3 Electroslag remelting (ESR)
      4. 1.6.3.4 Commercial thoughts
   4. 1.6.4 Shaped castings
7. 1.7 Potential for quality assurance

1.1. Theoretical strength of liquids

1.1.1. Classical continuum theory

When a liquid is subjected to an increasing hydrostatic tension, equivalent to a negative pressure, it will eventually “fracture.” The fracture process for liquids is the formation of a pore. Above a critical size it will expand explosively, relieving the elastic stress in the liquid. The critical stress in the liquid at which this occurs is called “the fracture pressure, P_f.”

It was probably Temperley in 1953 who first proposed that the tensile strength of liquids might be estimated from the minimum in the pressure P versus volume V relation predicted by the van der Waals's equation. As the van der Waals description derives from the kinetic theory of gases, being a first approximation of the behavior of a highly compressed gas, this is definitely a counterintuitive approach, but illustrates the power and delight of physics. The well-known general gas equation is

P V = NRT

N is Avogadro's number, being approximately 6 × 10²³; the number of atoms per mole. If we modify this relation introducing the constants “a” and “b” to take approximate account of the forces and volumes, respectively, of the atoms, we have the van der Waals equation

( P + a N 2 V 2 ) ( V − N b ) = NRT

It is noted that when Nb = V, the whole volume is occupied by atoms so that the pressure approaches infinity, but when V is large, the equation approaches the general gas equation. This equation is a cubic in terms of the volume V (as is clear if the terms of the equation are all multiplied by V²) and so has either 1, 2, or 3 solutions for V for a given value of P, as seen in Fig. 1.1.

The left-hand branch of the curve shows how the volume of the liquid will increase slightly as the pressure falls, even when pressure continues to fall, becoming negative, as a tensile stress (but a hydrostatic tensile stress of course means the stress has the same value in all directions). However, at P_f, pressure becomes a minimum, corresponding to the ultimate tensile strength of the liquid. Beyond this point, the pressure is predicted to jump unstably to the maximum vapor pressure. This upward-sloping region of the curve is unstable because the volume is predicted to increase at the same time as the pressure is increasing, which is clearly not realistic. For the right-hand portion of the van der Waals curve, the metal has become a vapor, which, with increasing pressure, reduces its volume as would be expected.

Figure 1.1 The van der Waals relation between pressure P and volume V of a liquid, indicating its theoretical strength.

The equation has to be solved iteratively to find P_f. We shall not attempt that here. Solutions are posted on the internet for interested readers. In addition, Fletcher (1993) gives an excellent account. We simply note that the equation predicts values for P_f approximately in agreement with those of simpler approaches described below.

1.1.2. Classical bubble nucleation

Using an alternative approach of classical physics, it is easy and quick to demonstrate that a pore cannot be nucleated in a liquid metal: the well-known equation for the mechanical stability of a spherical pore is

P = 2 γ / r

(1.1)

where P is the internal pressure (or external hydrostatic tensile stress), γ is the surface tension, and r is the radius of the pore. For liquid aluminum and liquid iron, we can take γ as approximately 1 and 2 N/m, respectively.

We do not know the radius r to find out P_f. However, we can guess it is only a very few atomic diameters. To obtain an estimate, if we assume Fisher's values (see below) for P_f are between approximately 3 and 4 GPa for Al and approximately 7.0 GPa for Fe, Eq. (1.1) indicates the critical bubble sizes are close to 2–2.5 atom diameters (assuming atom radii 0.28 and 0.25 nm, respectively) corresponding to a pore size of approximately 8–14 vacancies in each case. These values seem reasonable and are taken as approximately correct from now on.

1.1.3. Homogeneous nucleation

The above estimates for the ultimate strength of liquid metals are confirmed to within a factor of about 2 by a number of elegant, classical theoretical studies to compute the theoretical strength of liquid metals. The impressive compilation of data by Tiryakioglu for aluminum confirms the reliability of the strength data (Fig. 1.2). In the short review of this topic by the author (Campbell, 1968), the treatment by Fisher (1948) stands out because of its simplicity and elegant logic.

Fisher quantifies the conditions required for the formation of porosity in liquid metals. From energy considerations, using macroscopic concepts such as surface tension, he finds the quantity of work associated with the reversible formation of a bubble in a liquid. If the local pressure in the liquid is P _e , we need to carry out an amount of work P _e V to push back the liquid far enough to create a bubble of volume V.

Figure 1.2 The ultimate tensile strength of solid and liquid aluminum as a function of temperature (Tiryakioglu, 2018).

The formation and stretching o...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• Dedication
• Preface
• Acknowledgments
• Introduction
• Chapter 1. The fracture of liquids
• Chapter 2. Fracture in the liquid/solid state
• Chapter 3. Fracture of solids
• Appendices
• Postscript
• References
• Index