Steels: Microstructure and Properties

distance

Chapter 1

Iron and Its Interstitial Solutions

Abstract

Pure iron is remarkable in its complexity, not only because of its many allotropic forms. There are hidden features related to magnetism which, for example, make the expansion coefficient of austenite greater than that of the more loosely packed ferrite – one consequence of this is that austenitic steels deteriorate when subjected to a combination of stress and thermal fluctuations. We discuss here the choreography of atoms during solid-state phase transformations together with the role and behaviour of interstitial atoms such as carbon, nitrogen and hydrogen. The mobility of substitutional solutes is described to emphasise some counterintuitive observations such as the fact that heavy atoms like molybdenum actually diffuse faster than the iron in which they are dissolved.

Keywords

Allotropes of iron; Magnetic properties; Interstices; Transformation mechanisms; Diffusion; Interstitial solution

1.1 Introduction

Iron is created in the stars, which at first fuel themselves through the fusion of hydrogen into helium. When hydrogen is exhausted, the fusion of helium leads to the creation of carbon. This exothermic process of nuclear burning can, in sufficiently massive stars, lead to the formation of even heavier elements, the sequence ending with the creation of iron. Iron is the most stable element in the universe and further fusion to form even heavier elements does not release energy. As a consequence, the fuel supply becomes exhausted and the star collapses into states that depend on its mass. If the core becomes so heavy that it cannot support its own gravity, a giant explosion (a supernova) occurs that produces in a short time scale, most of the elements beyond iron.

Impressive as the cosmological origins of iron and carbon are, the story that truly is worth telling can be witnessed in every aspect of terrestrial life. Such is the success of iron, that steel forms the ‘gold-standard’ against which emerging materials or supreme acts of endeavour are compared. What is often not realised is that this is a moving standard, with notoriously regular and exciting discoveries being made in the context of iron and its alloys. This is why steel remains the most successful and cost-effective of all materials, with more than a billion tonnes being consumed annually in improving the quality of life. This book attempts to explain why steels continue to take this pre-eminent position, and examines in detail the phenomena whose exploitation enables the desired properties to be achieved.

One reason for the overwhelming dominance of steels is the endless variety of microstructures and properties that can be generated by solid-state transformation and processing. Therefore, in studying steels, it is useful to consider the behaviour of pure iron first, then the iron-carbon alloys, and finally the many complexities that arise when further solutes are added.

Pure iron is not an easy material to produce. It has nevertheless been made with a total impurity content less than 60 parts per million (ppm), of which 10 ppm is accounted for by non-metallic impurities such as carbon, oxygen, sulphur and phosphorus, with the remainder representing metallic impurities. Iron of this purity can be extremely weak when reasonably sized samples are tested: the resolved shear stress needed to induce slip in single crystal at room temperature can be as low as 10 MPa, while the yield stress of a polycrystalline sample at the same temperature can be well below 50 MPa. However, the shear strength of small single crystals has been observed to exceed 19,000 MPa when the size of the sample is reduced to about 2 μm. This is because the chances of finding crystal defects such as dislocations become small as the size of the crystal is reduced. The ideal tensile strength of a perfect crystal of iron pulled along 〈100〉 is about 14,200 MPa [1].

For comparison purposes the breaking strength of a very small carbon nanotube has been measured to be about 130,000 MPa; this number is so astonishing that it has led to misleading statements about their potential in structural applications. For example, the tubes are said to be a hundred times stronger than steel; in fact, there is no carbon tube which can match the strength of iron beyond a scale of 2 mm, because of the inevitable defects which arise as the tubes are grown [2,3].

The lesson from this is that systems which rely on perfection in order to achieve strength necessarily fail on scaling to engineering dimensions. Since perfection is thermodynamically impossible to achieve in large samples, steels must in practice be made stronger by other means which are insensitive to size. The mechanisms by which the strength can be increased will be discussed – suffice it to state here that it is possible to produce steel with a strength of 5500 MPa, with sufficient ductility to ensure safe application. Some of the methods by which such impressive combinations of properties are achieved without compromising safety will be discussed, before the wide range of complex structures which determine the properties is dealt with.

1.2 Allotropes of pure iron

At least three allotropes of iron occur naturally in bulk form, body-centred cubic (bcc, α, ferrite), face-centred cubic (fcc, γ, austenite) and hexagonal close-packed (hcp, ε). The phase β in the alphabetical sequence α, β, γis missing because the magnetic transition in ferrite was at one time incorrectly thought to be the β allotrope of iron, responsible for hardening when the iron is quenched. Nevertheless, it is true that α-iron is not strictly cubic in its ferromagnetic state below the Curie temperature [4]. This is because the magnetic spins are aligned, say along the z axis, so that rotations about the x or y axes must be combined with time reversal to preserve the directions of the spins. The magnetic point group becomes tetragonal. It follows that the ferrite structure below the Curie temperature is tetragonal, but the extent of tetragonality is very small and neglected in most experiments on steels containing complex microstructures where X-ray diffraction peaks are too broad to detect small differences in lattice parameters.

In fact, there are magnetic transitions in all of the allotropes of iron. The phase diagram for pure iron is illustrated in Fig. 1.1. Each point on any boundary between the phase fields represents an equilibrium state in which two phases can coexist. The triple point, where the three boundaries intersect, represents an equilibrium between all three phases which coexist. It is seen that in pure iron, the hcp form is stable only at very large pressures, consistent with its high density. The best comparison of the relative densities of the phases is made at the triple point where the allotropes are in equilibrium and where the sum of all the volume changes is zero:

. The core of the earth is predominantly iron, and consists of a solid inner core surrounded by a liquid outer core. Knowledge of the core is uncertain, but it has been suggested that the crystal structure of the solid core may be double hcp, although calculations which assume pure iron, indicate that the ε-iron remains the most stable under inner-core conditions.

Figure 1.1 The phase diagram for pure iron (data from Bundy [5]). The triple point temperature and pressure are 490∘C and 110 kbars, respectively. α, γ and ε refer to ferrite, austenite and ε-iron, respectively. δ is simply the higher temperature designation of α. However, below the Curie temperature, α-iron is slightly tetragonal with a difference in lattice parameters of just 6 × 10−15 m, determined from magnetostriction experiments [6].

1.2.1 Thin films and isolated particles

surface of a fcc substrate leads to trigonal iron.

Very thin films of iron retain their ferromagnetic character, but there are special effects due to the small dimensions. The magnetic moment per atom becomes very large: 3.1 Bohr magnetons compared with 2.2 for bulk α.

Many classical studies of nucleation theory have been conducted on minute (5–1000 nm) particles of iron where the defects responsible for heterogeneous nucleation can be avoided. Such particles have acquired new significance in that they are exploited in the manufacture of carbon nanotubes. The particles are deposited due to the decomposition of ferrocene in chemical mixtures which also contain the ingredients necessary to grow the tubes. Small particles of iron also have a role in increasing the thermal conductivity of fluids when in suspension [7].

It is expected that the coarser particles will have the bcc crystal structure of ferrite, but it has to be appreciated that a 5 nm particle has about half its atoms at the surface. Metal surfaces are prone to reconstruction into a variety of two-dimensional structures which will complicate the interpretation of the structure of the particle as a whole. The surface also plays another role, in that it alters the total free energy of the particle leading to a depression of its melting temperature. It has been estimated that a 5 nm diameter iron particle could melt at a temperature as low as 500∘C. Fig. 1.2 illustrates an iron particle inside a carbon nanotube – its bulbous character has been speculated to be due to melting.

Figure 1.2 A multi-walled carbon nanotube containing a particle of iron (unpublished micrograph courtesy of I. Kinloch).

Small metal particles in the size range 1–5 nm are close to a metal/insulator transition. When observed at the tips of carbon nanotubes using scanning electron microscopy, the iron particles have shown a tendency to charge, possibly indicating a loss of metallic behaviour.

1.3 Austenite to ferrite transformation

The vast majority of steels rely on just two allotropes, α and γpoint). This high-temperature ferrite is traditionally labelled δ, although it is no different in crystal structure from α. The δ-ferrite remains the stable phase until melting occurs at 1536∘C.

transformation is accompanied by an atomic volume change of approximately 1–3%, which can lead to the generation of internal stresses during transformation.

Figure 1.3 Temperature dependence of the volume per mole of atoms in iron crystals (data adapted from [8]).

The detailed geometry of unit cells of α- and γ-iron crystals is particularly relevant to, e.g., the solubility in the two phases of non-metallic elements such as carbon and nitrogen, the diffusivity of alloying elements at elevated temperatures and the general behaviour on plastic deformation. The bcc structure of α-iron is more loosely packed than that of fcc γ-iron (Fig. 1.4). The largest cavities in the bcc structure are the tetrahedral holes existing between two edge and two central atoms in the structure, which together form a tetrahedron. The second largest are the octahedral holes which occupy the centres of the faces and the 〈001〉 edges of the body-centred cube. The surrounding iron atoms are at the corners of a flattened octahedron. It is interesting that the fcc structure, although more closely packed, has larger holes than the bcc structure. These holes are at the centres of the cube edges, and are surrounded by six atoms in the form of an octagon, so they are referred to as octahedral holes. There are also smaller tetrahedral interstices. The largest sizes of spheres which will enter these interstices are given in Table 1.1.

Figure 1.4 Schematic representation of the octahedral and tetrahedral interstices in ferrite ( α ), austenite ( γ ) and hexagonal iron ( ε ). The filled circles represent the positions of iron atoms whilst the interstices are represented by open circles together with their coordination polyhedra. Adapted from [9].

Table 1.1

Size of largest spheres fitting interstices in bcc and fcc iron. r = atomic radius of iron, N is the number of interstices per iron atom. Typical coordinates of the interstices are expressed as fractions of the cell parameters. The calculations assume hard sphere models for the crystal structures

transformation in pure iron occurs very rapidly, so it is not generally possible to retain the high-temperature fcc form at room temperature. Rapid quenching can substantially alter the morphology of the resulting α-iron, but it still retains its bcc structure. It follows that any detailed study of austenite in pure iron must be done at elevated temperatures, e.g. using X-ray or neutron diffraction. The transformation of the austenite on cooling can also be followed using diffraction based on the intense X-rays generated in a synchrotron, or using precision dilatometry. The latter technique relies on the volume change accompanying the transformation from austenite to ferrite.

, depending on its density).

1.3.1 Mechanisms of transformation

One of the reasons why there is a great number of microstructures in steels is because the atoms can move in a variety of ways to achieve the same allotropic transition. The transformation can occur either by breaking all the bonds and rearranging the atoms into an alternative pattern (reconstructive transformation), or by homogeneously deforming the original pattern into a new crystal structure, i.e. displacive or shear transformation, Fig. 1.5.

Figure 1.5 Schematic illustration of the mechanisms of transformation. The parent crystal contains two kinds of atoms. The figures on the right represent partially transformed samples with the parent and product unit cells outlined in bold.

In the displacive mechanism the change in crystal structure also alters the macroscopic shape of the sample when the latter is not constrained. The shape deformation during constrained transformation is accommodated by a combination of elastic and plastic strains in the surrounding matrix. The product phase grows in the form of thin plates to minimise the strains. The atoms are displaced into their new positions in a coordinated motion. Displacive transformations can, therefore, occur at temperatures where diffusion is inconceivable within the time scale of the experiment. Some solutes may be forced into the product phase, a phenomenon known as solute trapping.¹ Both the trapping of atoms and the strains make displacive transformations less favourable from a thermodynamic point of view.

It is the diffusion of atoms that leads to the new crystal structure during a reconstructive transformation. Imagine that the transformation proceeds as by the displacive mechanism (Fig. 1.6a, b), but that the resulting shape deformation is eliminated by transporting the segment as in Fig. 1.6c to recover the overall shape shown in (d). This transport is the diffusion that is needed in order that the strain energy term is essentially eliminated as if there is fluid flow in the surrounding matrix. The flow of matter is sufficient to avoid any shear components of the shape deformation, leaving only the effects of volume change. In alloys, the diffusion process may also lead to the redistribution of solutes between the phases in a manner consistent with a reduction in the overall free energy.

Figure 1.6 Phenomenological interpretation of reconstructive transformation. The virtual operation that eliminates the shape change is the diffusion required during a reconstructive transformation, irrespective of whether it occurs in pure iron or in an alloyed steel [10].

It should be emphasised that the diffusional flow is essential even in pure iron, where the reconstructive ferrite and martensite are distinguished purely by the consequences of the shape deformation in the latter case. transformation by a displacive mechanism [11]. All the phase transformations in steels can be discussed in the context of these two mechanisms (Fig. 1.8). The details are presented in subsequent chapters.

Figure 1.7 A whisker of iron, originally straight, transformed partially to austenite – the image is a frame from a movie taken to show the shape deformation due to the unconstrained α  →  γ transformation. The position of the kink represents the interface between the α and the striated γ . After Zerwech and Wayman [11], reproduced with permission of Elsevier.

Figure 1.8 A selection of the structures generated by the decomposition of austenite. The term ‘paraequilibrium’ refers to the case where carbon partitions but the substitutional atoms do not do so. The substitutional solute to iron atom ratio is therefore unchanged by transformation.

1.4 Carbon, nitrogen and hydrogen in solution

1.4.1 Solubility in α- and γ-iron

It is often said that ‘ferrite has a solubility for carbon of about 0.02 wt%’. However, solubility is defined by equilibrium between two or more phases. The solubility of carbon in ferrite that is in equilibrium with austenite will therefore not be the same as when it is in equilibrium with cementite, Fig. 1.9. Ferrite which is in equilibrium with cementite has a particularly low solubility for carbon at low temperatures; Fig. 1.9c shows the precipitation of cementite occurring in a steel containing just 0.02 wt% of carbon, when the alloy is aged at 240∘C. The solubility will depend also on external parameters such as pressure, temperature, magnetic fields and stress.