Engineering Physical Metallurgy by Y. Lakhtin

arti ngineering hysical Gen ats 2 je) ae? ee 2) E> cas ‘Mir Publishers ===" CBS Publishers & Distributor: MOSCOW INDIAmt Y. LAKHTIN © ENGINEERING PHYSICAL METALLURGY Translated from the Russian Nicholas Weinstein Mir Publishers Moscow aD ons CBS PUBLISHERS & DISTRIBUTORS 4508/1 A, 11-DARYAGANJ, NEW DELHI-110U::2First published 1963 Second printing 1966 | Third printing 1968 | Fourth printing 1971 Fifth printing 1975 | Sixth peinting 1977 Ha anesilicron soune ISBN : 81-239-0602-1 First Indian Edition ; 1998 Copyright © 1983 by Butterworth Heinemann This edition has been published in India by arrangement with Butterworth Heinemann, U.K. Ail right reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system without permission, in writing, from the publisher. This reprinted edition is authorized for sale in India only. Published by S.K. Jain for CBS Publishers & Distributors, 4596/1A, 11 Darya Ganj, New Dethi - 110 002 (India). Printed at : Asia Printograph, Shahdaru, Delhi - 32CONTENTS Chapter 1. Geystal Structure of Motals . I. Aomic Structure of Metals. 13 Grystallerapie Nolaiion ‘of’ Atomic janes 1-4, The Structures of Actual Metal Crystals = 1-5. Properties of Metals... . Chapter 2. Soliditication and Metal Ingot Structure... 1. 1 es 2-1, Solldificetion of Metals. 2... 2-2, Metal Ingot Structure ....... Chapter $. Plastic Deformation and Beerttalsation In Metals... 0a. + 3-1. Plastic Deformation . + %2. Recovery and Recrystallisation ; oe Guapter mated ee Semeving Metal Strvctare «+ + Te ieeregiate (iacroera oe wae . 4-2, Microstructure a] een eee . 4-3. X-Ray Anal: me ae ee eee 4. Physical “tor ‘Studying AL Chapter 5. The Mechanical Properties of Metals... . « 51. Mechanical Testing of Metals . 5-2. Tension Tests . . . Chapter 6. Binary awe. Constitution and Equilibrium Diagrams . .. . 1. General Principies of Phase Transformations In Alloys . . 62. The Phase Rute and Equilibrium Diagrams .. 4... + SSS 8BRextss 2 geese 286 HRB BContents 6-3. Equilibrium Diagram of a Binary System in Which the Compo- nents Form a Mechanical Mixture of Crystals in the Solid State and Are Completely Mutually Soluble in the Liquid State... . 6-4. Equilibrium Diagram of a System Whose Components Are Mu- tually Soluble in Both the Sy quid and Solid States... . 65. Equilibrium Diagram of em in Which the Components Have Unlimited Solubility in the Liquid State and Form Chemical Compounds Upon Solidification... . 2... 2. ee ee 6-6. Equilibrium Diagram of a System Whose ‘Components Are Sub- ject to Allotropic Transformations .. 2. .....2 20+ e 6-7. Diagrams of Ternary Systems... .....0000005 5 Chapter 7. The Iron-Carbon Equilibrlum Diagram... ....... Tl Tron. ww ee See ee ee 7-2. The Iron-Carbon “Equilibrium Diagram . . Chapter 8. Phase Transtormations in the Iron-Carbon System... . . . #1. ree of Austenite (Transformations That Occur in n Heating Stee! see : 82. Austenite Grain ‘Growth jn Heating 222 I 8-3. Transformation of Austenite into % arts (Isothermal Decom- sition of Austenite). 2... ee an &4. Transformation of Austenite™ Upon’ Continuous Cooling 2. 8-5. Martensitic Transformation in Steel... .......-- 8-6. Te.apering of Steel... ee ee ee ee B7 Swing Sk eee eee C!-oter J, Vieat Treatment of Steel... ee ee eee 9-1, Annealing of Steel . 9.2. Normalising of Steel $3. Hardening of Stee. 94. ring of Steel PEPPiiibiiiiiiiiii: SF Sap bee treatment ot ‘Steel ee ee See ees 9-6. Defects Due to Heat Treatment of Steels. 2... . 0.00. Chapter 10. Surtnce Hardening of Steel... 2... 6 eee eee 10-1, High-Frequency Induction Hardening... . wee 10-2, Hardening with Electric Contact Resistance Fiesting eee 10-3. Hardening with Electrolytic Heating . ween 10-4. Oxyacetylene Flame Hardening... .....-.. 2.206. Chapter 11. Chemical Heat Treatment of Steel (Case-Hardening) . . . . 11-1, Physical nt le Anvolved in Chemical Heat Treatment . . . 11-2, Carburising of be ee . 11-3. Nitriding of Steel 2. 0... ee ee ee ae 1-4, Cyaniding and ee sbonttiiding “of Stel S251: nee 1-5, Dilfusion Coatings. 2... ee ee Chopter 12. Minor Constituents and Alloying Elements in Steel... . 12-1. Effects Produced by the Minor Constituents... ...... 22. Distributton of Alloying Elements in Steel . . Btlects of the Alloying Elements on Phase Transformations in 12-4, Structural Classes ‘of Alloy Steels |Contents Chapter 13. Steel 2... ee ee ee 13-1. General Classification .. 306 13-2, Ordinary and Improved Carl 308 13-3. Quality Carbon Structural Steels 310 13-4. Carbon Steels for Castings . 314 13-5. Free Cutting Steels 315 13-6, Low-Alloy Construct: 316 13-7, Alloy Structural Steels 318 13-8. Tool Steels + 333 13-9. Wear-Resistant (Ai ic) Steels . 13-10, Stainless and Acid-Resistant Steel 13-11, Scale-Resistant and Heat-Resistant Steels 13-12. Magnetic Steels and Alloys 13-13, Alloys with Definite Expansion and Elasticity Chapter 14. Cast Iron and Ita Heat Treatment + Wael. Grey Cast Iron. . . 369 142, Heat Treatment of Cast fron 377 14-3. Malleable Cast tron... 2... 379 Chapter 15. Copper and Its Alloys... 1 eee te ee ee ee 15-1, Copper... e 15-2. Brasses . 2 4 +e 15-3. Bronzes . . . 382 382 385. 389 Chapter 16. Nickel and Ite Alloys. 2... . eee ee ee ee ee 888 400 400 402 402 404 Chapter 17. Aluminium and Us Alloys . 17-1, Aluminium . 17-2. Classification of Aluminium * Alloys . 17-3, Non-Heat-Treatable Wrought Aluminium stable at some higher temperatures; y (gamma) at still higher temperatures; etc. The allotropy of iron is of especial practical importance. Iron exists in two allotropic forms: a-iron, with a body-centred cubic lattice, stable at temperatures up to 910°C and over 1,401°C, and § he coir wage EY 2 fee “ose-pacnt H 9 Gene pari feagonad Hg © Lam structures 6: <5 ann 3 g, f} of . Ss 9 4 g 83 . g > ge & 2 L g s gs, 1 : TN iil aa lcal ¥ JAs HOY Wol Vasaginlsat 21a Ce We i yyw. . o if (5é A Tees ics] be We Retr MATE BL =v ox sae os: 0 fo 20 30 40 7D bo 90 “30 60 Atomic number of the element Fig. 7. Interatomic distances in crysiais y-iron, with 4 face-centred cubic lattice, stable in the range from 910° to 1,401°C, There are also three allotropic modifications of manganese (@-Mn, B-Mn, and y-Mn) with complex crystai lattices, two modifications of cobalt (a-Co and B-Co), twa-—of tin (2-Sn and B-Sn), two~-of titanium (a-Ti and B-Ti), two-cf zirconium (B-Zr and a@-Zr), and two—of tel- lurium (a-Te and B-Te). Changes in the packing density of the crystal lattice, in conversion from one altotropic form to another, lead to changes in the volume of the material as well. For example, the packing of y-iron (C12) is 1.05 per ceat denser than that of a-iron (C8) and its specific volume is correspondingly18 x Crystal Structure of Metals less. This is of great importance, as we shall see further on, in the heat treatment of steel. In the transformation of white tin (B-Sn) to grey tin («-Sn), the volume increases by 26 per cent. This increase in volume hinders the allotropic ‘transformation B-Snzta-Sn, which takes place only on the surface of the metal. On the other hand, it causes asymmetric stresses in the metal which results in cracking of the brittle a-Sn and in its transformation into a grey crystallic powder (tin pest or tin plague). The high brittleness of a-Sn is a result of its less-closely packed crystallic structure (see Fig. 4, a, typical of nonmetals. 1-3. Crystallographic Notation of Atomic Planes The positions of atomic planes (planes passing through atoms) in crystal space lattices are determined by indices (A, & & which are three whole rational values, reciprocals of the intercepts. The Shoot c Fig. 8. Atomic planes in a body- centred cubic lattice: amcubic plane (100), b—actahedsal oom hedrat v ri plane wie vs diamand) dodeca edral Intercepts of a given plane are the distances from the origin of the cuordinate axes at which the plane intersects each axis. The units of length along these axes are equal to the lengths of the edges of the unit cell. | As a first example in designating such planes we may establish the indices of the planes in a cube. It is evident from Fig. 8 thatThe Structures of Actual Metal Crystals 19 each plane of a cube intersects only one axis and the intercepts will equal (1, 00, cv), (00, 1, 00), and (00, oo, 1}.* The reciprocals of the intercepts will respectively equal 1, 0, 0; 0, 1, 0; and 0, 0, 1. The indices of planes are usually enclosed in parentheses and are not separated by commas, as (100), (010), and (001). In addition to the planes of the cube (Fig. 8, a), a cubic lattice has octahedral planes (111) (Fig. 8,8) and diamond dodecahedral planes (110) (Fig. 8, c). It should be understood that the indices refer not only to one plane, but to whole sets of parallel planes. It is not difficult to see that the density in the arrangement of atoms may vary in different planes. In a body-centred cubic lattice, for example, the plane (100) has but one atom ("/,x4). The diamond dodecahedral piane (110) of the same lattice has two atoms; one from the atoms at the corners (#/,X 4) and one in the centre of the cube. The plane with most closely packed atoms of a face-centred cubic lattice will be the octahedral plane (111); in a body-centred cubic lattice. it will be plane (110). 1-4. The Structures of Actual Metal Crystals Wide application of the electron microscope and X-ray analysis for investigations of the metal structure made it possible to establish the fact that no actual single crystal (i.e., monocrystal) or even any grain of a polycrystailine metal has a perfectly regular arrangement of atoms throughout its volume. Each actual crystal consists of separate blocks, often called blocks of mosaic structure, from several microns to hundredths of a micron in size, each tipped slightly out of alignment, through an angle of several minutes, with its neighbours. A diagram illustrating mosaic structure is given in Fig. 9, a while Fig. 9, 6 shows a real stiucture as. observed with the aid of an electron microscope. The reasens for the occurrence of mosaic structures have not yet been completely expla are undoubtedly associated with a special type of cutis Chapter 2). A study of mosaic fbiock} siructure ls of great importatice since the size of the blocks considerabiy aifects the properties of the metal. Unde: ordinary conditions, a metal consists of a great number ot randarty oriented grains (crystallit: usually sf is egular form (see Fig. 16). On the boundariss of the graine, the atoms Fave a much less regular arrangement than in the giain body, due in the influence of atomic iorce fields in the neighbouring grains ( . It should be Noied also that in engineering metals, all sorts of immvurites are con- * A plane parailel to an axis docs net intersect it at atl and the iniccrept is bis ted as 0; i, e, if is infintiely great. *2 Crystal Structure of Metals centrated on the grain boundaries and they increase misorientation in atomic arrangement. A certain degree of misorientation (or distortion) probably exists on the boundaries of mosaic blocks as well, The atoms in the boun- Fig. 9. Mosaic structure: a—schematle diagram, b--syucture of aluminium, x 700 dary layers will possess higher energy due to distorted packing (Fig. 10). In exactly the same manner, atoms located on the surface of the crystal possess increased energy due to uncompensated forces of atomic interaction. érain I _feutidary bayer Greindy 9696698650 9, 990568 e999 Soeoee 8990600 60686 eoooes PPPP ED roblay 1 alata Fig. 10. Model illustrating atomic arrangement in the boundary layer between two crystals and the variation in the potential energy of the atoms (affer K. P, Bunin) This surplus energy, known as surface energy, is of prime impot- tance in phase transformations of metals and alloys. It is incorrect to consider each atom as being firmly bound to a definite point in the crystal lattice. Diffusion takes place in crystallic bodies, including self-diffusion, i.e., the rearrangement of atoms ofThe Structures of Actual Metal Crystals a a single material within its crystal lattice. The process of diffusion in a crystallic body is explained by the fact that certain atoms, whose kinetic energy substantially exceeds the average value due to the effect of thermal vibrations, may leave their regular positions at the points of the crystal lattice (Fig. 11, a) and jump into an interstitial position. An atom which has left its lattice site is said to be dislocated. This movement from a lattice point to an_ interstitial position will create a vacant lattice point (or vacancy) which may be occupied by any other atom (Fig. 11, 6). Migration of atoms Interstitial atom tothe surrace — cattice vacaney tdistocated} Fig. 11. Diagrams illustrating point imperfections of a crystal (a and 6) and diffusive migration of atoms (holes) (b) The higher the temperature, the more dislocated atoms and vacancies there will be. However, even at temperatures approachin the melting point their number will not exceed 2 per cent of the total amount. Atoms surrounding a dislocated atom are naturally spread apart somewhat, since the diameter of an atom is always larger than the interstitial spaces. Atoms surrounding a vacant lattice point, on the other hand, are located somewhat closer. Vacant lattice points in a crystal do not remain stationary. A vacancy may migrate in the process of thermal vibrations since, sooner or later, one of the atoms surrounding the vacancy will jump into it (for example / in Fig. 11,6). This frees another lattice point which will be taken up by 2, etc. Thus, the vacancy continuously migrates in the crystal, "Movement of atoms to interstitial spaces, formation of vacancies and their migration, and distortion of the lattice around dislocated2 Crystal Structure of Metals _ atoms and vacancies are reasons for departures from geometrically perfect structures of crystals. Distortions of the crystal lattice may also occur as a result of the in- corporation of foreign atoms of the impurities which, as a rule, are present in even the “purest” metal. For example, monocrystals of germanium can now be obtained with a purity of 99.99999999 per cent (one atom of an impurity per ten bil- lion atoms), but even in this case one gram-atom of germanium will contain 10'* foreign atoms. The lattice around the impurities is distorted (Fig. 11, a); its structure is impaired and, consequently, higher energy is peculiar to the atoms as a whole in the parts of the lattice surrounding foreign atoms. So-called dislocations can also be formed in the crystal. These likewise impair the regularity of the crystal lattice. Fig. 12. Regions of positive (|) and negative (J) dislocations Shown in Fig. 12 are so-called edge dislocations. An edge dislocation is 8 localised distortion of the crystal lattice due to the presence cla an extra?’ atomic plane (PQ or P’ @’), perpendicular to the plane of rawing. This extra plane (see Fig. 12, 6) may be either above (PQ) or below (P’ Q’) the slip plane (shown by the dash line AC). In the former case, the dislocation (PQ) is said to be positive and is denoted by the symbol 1; in the latter case, it (P’ Q’) is said to be negative and is denoted by the symbol T. - The edge of an extra plane is a line of the edge dislocation whose I may Teach many thousands of interatomic distances. The crystal lattice is severely distorted in a region extending over several ato- mle distances from a dislocation. Distortion rapidly disappears beyond Dislocations have been observed directly by means. of an electron microscope. - Dislocations are formed in the process of solidification of metals and mainly in their deformation.Pi Metals 23 : The dislocation density * in a metal prior to deformation ranges from 10‘ to 10* per cm~*; after deformation it may reach 107* per cm=*. New dislocations occur in the Pipers of plastic deformation. In this case, the increased number of dislocations is created by the so- calied Frank-Read sources (or Frank-Read dislocation generators). Adescription of this mechanism may be found in special literature ** and is not within the scope of this book. 1-5. Properties of Metals Certain typical properties characterise metals. These properties are associated with the internal structure of metal Is. The first scientific definition of the word “metal”, formulated by the great Russian scientist, Mikhail Vasilyevich Lomonosov—“A metal is a bright solid that can be forged"—still holds true in our times. Metallic lustre and the ability for plastic deformation are, in fact, characteristic features of metais. The most important feature, however, of typical metals is their high etectrical conductivity of the electron type, which decreases as the temperature is raised. Many properties of metal crystals (physical, mechanical, and others) often depend on the direction along which they are measured. All crystals are anisotropic, as differing from isotropic amorphous solids Glos, plastics, and others which display the same properties in every plane and every direction). For example, the modulus of elasticity of a monocrystal of zinc has a value of 3,560 kg per sq mm in a direction parallel with its main axis and a value of 12,630 kg/sq mm in a perpendicular direction. In the same manner, the thermal coefficient of -fineer expansion (a=7x$}) varies from 68.9x10-* to 14.1% x10-* mm/mm-degree C. ” Properties of a monocrystal are based on the different arr: ents of the stoms in various directions. The arrangement of the atoms of an amorphous solid is more or less equally irregular in all directions and, therefore, the properties do not have a directional nature. * The dislocation density is e=% om-%, where Zi Is the total length of the dislocations in the crystal, in cm; V is the volume of the crystal, in cm*. ba At the present time, the theory ot dislocations is being satensively used to er) in phenomena occurri in metal in particular, success- f ly used in the study of the cesses of diffusion, plastic deformation, mechanical eae {oaei crow ete. A detailed ifter cold heat resistance, jon. of the theory of ate ie eens following special tre- tons and Sie We T. Read, Jes Dislocations in McGraw-Hill Book Company, Inc. New York, 1963; A. H. Cottrell, Disloca Plastic Flow in Crystals, Oxtord University Press, New York, 1963; and A. J. Forty, “Direct Observations of Dis- locations in 1”, Advances in Physics (A quarterly supplement of the Philo *sophical Mi » Jan. 8, 1064, No. 9}.4 Crystal Structure of Metals Since actual or engineering metals are solids consisting of a great number of crystals (i.e., polycrystalline solids) with anisotropic properties, the question naturally arises as to whether the metal as a whole is anisotropic in respect to its properties. In cases where the crystallites have the same orientation, the polycrystalline solid they compose will also be anisotropic. If the crystallites are differently oriented relative to each other, the summary properties of all the crystals will be approximately the same in all directions. This apparent isotropy will be retained only until the crystals are reoriented along a single direction by some technological process, for example, considerable cold working. After this, the metal will again be anisotropic.Chapter 2 : SOLIDIFICATION AND METAL INGOT STRUCTURE 2-1. Solidification of Metals Transformation of a metal from the liquid to the solid state proceeds due to the conversion of matter to a more stable thermodynamic condition with less free energy F. If the transformation is accompanied by only a small change in volume, then * F=H-TS in which H is the total energy (enthalpy) of the system, T is the absolute temperature, and S is the entropy. At a given temperature, the total energy content (H) of a system can be considered to be composed of two portions: the free energy F and the bound energy TS. Both free and bound energy may occur simultaneously in several differerit forms within the system, but again all of these various forms can ultimately be converted into heat. For this reason, both types of energy are usually evaluated in thermal units, their sum being referred to as the heat content of the system. Free energy (F) is best described as usable energy in isothermal processes; in other words, under certain conditions it can be released by and removed from the system with no accompanying change of temperature in the system. Bound energy is not usable in isothermal processes; that is, it cannot ordinarily be liberated from a system without changing the temperature of the system. Bound energy, whose magnitude is expressed by the product TS, exists principally as sensible heat, i.e., as kinetic energy of the random motions of atoms within the system and as potential energy of their arrangements relative to each other. Fig. 13 schematically represents the variation in free energy of the liquid and solid states in relation to the temperature. At a temperature of 7',, the magnitudes of the free * According to the Second Law of Thermodynamics, not a single process will proceed spontaneously without a decrease in the free energy of the system concerned,% Solidification and Metal Ingot. Structure energy for the two states are equal. This temperature is the equilibrium temperature of freezing (or melting) for the given material at which both phases (liquid and solid) may exist simultaneously. At temperatures above T,, liquid metal will be more stable, having less free energy; at lower temperatures, solid metal will be more.stable. For the: development of the solidification process (freezing), there must be a difference in free energy which is due to the lesser free energy of solid metal in comparison with the liquid state. Therefore, the process of solidification will proceed only if the metal is supercooled below the equilibrium temperature of solidification 7,. The difference between the equilibrium temperature of solidification T, and the temperature T, at which this process proceeds under given conditions is called the degree of supercooling AT=T,—T, The higher the rate of cooling, the larger the degree of supercooling’ will be, since Fig. 3, Free energy vs temperature for the 7 Will be correspondingly liquid and solid states . Thermal «curves, characte- rising the solidification ‘process in pure metals and obtained at. various cooling rates, are given in Fig. 14. At very low rates of cooling the degree of supercooling is small and the solidification process proceed sata temperature near to the equilibrium temperature (Fig. 14, curve f). The horizontal section of the curves (representing a stop in temperature fall) is due to the evolution of the latent heat of fusion which compensates the heat dissipated to the surrounding atmosphere. The degree of supetcooling increases with an increase in the rate of cooling (see Fig. 14, curves 2, 3, # and 5) and the solidification process roceeds at temperatures substantially below the equilibrium jem ure of freezing. e maximum de of supercooling depends on the nature and purity of-the metal but usually it does not exceed 30°C. The purer 8 liquid metal is, the higher its stability will be and the more it is subject to supercooling.Solidification of Metals if » — In studying the structure of steel ingots, the great Russian metal- *; lurgist Dmitri Konstantinovich Chernov first established that the Time -———— Fig. 14. Cooting curves of a pure metal Fig, 1. Solidification of a metal (schematic) (alter I. L. Mirkin) J. process of solidification begins with the formation of nuclei (embryos) . of centres of crystallisation and proceeds with their growth. : -A schematic picture of the process of solidification. is shown in Fig. 15. Upon supercooling of the liquid metal below 7,, separate crystals form, as shown in the diagram, and begin to grow. As long28 Solidification and Metal Ingot Structure as the crystals grow freely, their geometry may be considered to be almost perfect, but as soon as the growing crystals mutually obstruct each other, their regular form is violated. At the places where they interfere with each other, further growth of the faces ceases; it continues only in those directions in which free liquid is still accessible. As a result of this process, the growing crystais, having a regular internal arrangement of atoms, take on an irregular shape after solidifying. They are called crystallites or grains (Fig. 16). Fig. 16. Structure of a polycrystalline metal Phenomena associated with the process of solidification are aomplex and varied. It is especially difficult to conceive the initial stages of the process when the first crystals or centres of crystallisation appear. Evidently, to explain the appearance of these centres it will be essential to understand the structure of the initial (matrix) phase,, i the liquid metal. Jt has been established at the present time that in a liquid metal, as distinct from the vaporous state where atoms are chaotically arranged, a certain “order” may be observed in atomic arrangement, not- withstanding the high mobility of the atoms, due to the forces of inte- taction between them. The atoms (ions) in a liquid metal do not move about freely. As in a crystalline metal, they oscillate about certain positions of equilibrium. In a crystalline solid, however, the atoms oscillate about certain mean positions of equilibrium that are regularly located in space and maintain these positions over sufficiently long periods of time; while in a liquid metal these mean positions, themselves, are in continuous motion. X-ray research has shown that at temperatures near to solidification, groups of atoms may come together in small volumes of the liquid metal; they have an atomic arrangement near to that in the solid metal. These groups or clusters may appear, exist for a certain time and then fall apart to appear again at some other place. Under definite conditions, however, such clusters may acquire higher-stability and become centres of crystallisation or nuclei.Solidification of Metals 9 The formation of nuclei is facilitated by unequa! distribution of energy between the atoms of a substance. At each given temperature, most of the atoms have an energy equal to a certain average value. In small volumes of substance, however, there are always a certain number of atoms possessing energy of a quantity either smaller or larger than the average value, These random and tentative deviations in the energy of separate atoms or regions of atoms from the average value for the given temperature are called fluctuations of energy. Correspondingly, the temperature, which determines the thermal energy of the matter, will vary over a wide range in these regions. As a result of the aforesaid, nuclei, capable of subsequent growth, will appear in certain small volumes of the liquid having a temperature lower than the mean value. Numerous investigations have proved that nuclei of a great variety of sizes may appear in freezing. Not all nuclei, however, are capable of further growth. The formation of nuclei changes the free energy of the system (Fig. 17). The appearance of a nucleus and its growth as a crystal will reduce the amount of free energy per unit volume (Fig. 17, curve /) due to the lesser free energy of a unit volume of a solid in comparison to the liquid state. At the same time, the appearance of an interface between the nucleus and the liquid medium causes an increase in the free energy of the system (Fig. 17, curve 3} since atoms on the interface possess higher potential energy. Therefore, a certain amount of energy is expended to form this interface between the phases. This additional energy is equal to the product of the nucleus surface area and the surface tension. The change in total free energy due to the formation of the nucleus and the phase interface is represented by curve 2 in Fig. 17 *. Ht is evident from curve 2 of Fig. 17 that the growth of a nucleus having a radius less than R, is impossible since this is accompanied by an increase in free energy. In nuclei with radii Jess than R,, the ratio of the surface area to the volume is large. The increase in free energy, associated with the forraation of the interface, is greater than the reduction in free energy due to the formation of a nucleus. This increases the free energy of the system. ot a nucleus appears with a radius exceeding R,, it will be stable and capable of further growth since an increase in its size will reduce the free energy of the system (Fig. 17, curve 2). The minimum size of * The variation in free energy of a system AF, associated with the appearance of a solid phase nucleus, is expressed by the equation AF =-— V(F,—Fg)- So, in which V is the volume of the nucleus, S is its surface area, o is the coefficient of surface tension between the liquid and solid phases, and Fs and Fy, are the free ener- gies per unit volume of the solid and liquid phases, respectively.0 Solidification and Metal ingot Structure nucleus (R.}, ible of growth: at a given temperature, is‘called the critical size of the nucleus.* Additional ‘energy, however, equal to */2 0 S,, is required for the formation of critical nuclei. This energy is provided: by’ the system and is due to energy fluctuations. Thermodynamic calculations show that the size of the critical nucleus depends on the degree of supercooling. The higher the degree of supercooling, i.e., the lower the temperature at which freezing proceeds, the smaller the size of the critical nucleus . will be (Fig. 18). The formation of nuclei in the liquid metal or nucleation, in a lance with the above-described mechanism, is called Vertatton in free energg Stza of nustens (2) size gf nucleus (A Fig. 17, Variation of the free energy of Fig. 18. Variation In the free energy of a a metal upon the formation of a nu- metal versus nucieus size for various - cleus of superoooling spontaneous nucleation. Nucleation is much facilitated if sotid particles are it in the liquid metal (walls of the ingot mould, various in- clus’ ete.) which serve as ready-made substrates for the formation of centres of crystallisation. : This is explained by the fact that the work required to form a nucleus on an existing surface is substantially less than that required for spontaneous nucleation in the liquid metal, Jt must be noted, however, that foreign inclusions will facilitate nucleation only if the suriace tension betweea the nuclei and the inclusion is less than that at the interface between the liquid phase and the solid crystals. It has been established that the nearer the crystallographic structure of the nuclei being formed is to that of the solid particles in the liquid metal, the less the surface tension at their interiaces. ° A ccitical nucleus (with a radius R,) may elther grow or melt again.‘Solidification of Metals at The formation of centres of crystallisation on such ready-made rates complies with the principle of structural and size con- ity. The nucleus forming on a ready-made substrate is oriented that the'atoms on its external face are arranged, in their configu- ion, in the same manner as those of the contacting face. The parame-. Mers of the lattices concerned should differ but slightly. = growth of a formed nucleus, usually called a three-dimenstonal a s, proceeds by building up new single-atom layers on its surface. : are called two-dimensional nuclei and they must be of a size not fess than a definite critical value, The addition of single atoms or yroups of atoms, less than critical in size, to the ing nucleus is panied by an increase in the free energy of the system and the ‘get atoms will be transformed back to the liquid phase. - The formation of a two-dimensional step (i.e, layer of single- , atons thickness) takes place in the same way as that of three-dimensio- nai nuclei. . wh Ma increased degrees of supercooling, the critical size of a two- dimensional nucleus and, consequently, the fluctuations of energy re- fired’ for its formation, are *_eThe tate of solidification de. Ff :-(peads on: ~""s4) the rate of nucleation, ’-dLe., upon the number of three dimensional nuclel appesring unit time in a unit volume A seamaz) 8142) the linear rate of crystal th, i.e., upon . tte “rate of increase in the tnear dimensions of crystals . per unit time (mm per sec). i: The linear rate of crystal ) owth (RO and the rate of Je Degree of sepercaating (aT) “mucleation (RW) both depend 47°98 ; onthe degree of supercooling Fig. 19. Emect of the degree of su ‘SOG- Oe. 19). fing on the rates. of nucleation (RM) end {Wels evident from Fig. 19 that crystal growth (RO) at the equilibrium temperature ‘2 RG and RN equal zero and, therefore, no freezing takes 4 Upon an increase in the d of supercooling both RG and RN ‘and reach a maximum at a definite rate of supercooling and * then drop again. The maximum for RG Is always reached at less su- pescooling than that for RW. : -Ancreased rate of solidification, due to an increased degree of su- ling, may be explained by the reduction in the critical size of t32 Solidification and Metal Ingot Structure the nucleus and, consequently, by less work required for its formation *. At high degrees of supercooling, however, the mobility of the atoms is reduced by the increase in viscosity of the liquid phase. This impedes nucleation and reduces RG and RW (see Fig. 19). Only the rising sections of RG and RN curves (solid lines in Fig. 19) have a practical significance for metals. Metals solidify sooner than degrees of supercooling are obtained which cause reduction of RG an : The larger the number of nuclei and the slower their growth, the smaller the crystals will be that grow out of each nucleus (i.e., the smaller the metal grains). The number of grains (n) (and, consequently, their size) may be computed from the following equation as a function of RG and RN: VB Since RN increases faster than RG for higher degrees of supercooling (Fig. 19), the higher the degree of supercooling, the smaller the grains will be. As a very high degree of supercooling cannot be achieved in liquid metal, smaller grain size is obtained in actual practice, at the present time, not by increasing the rate of cooling, but by introducing dispersion particles (inoculation) into the metal. These particles either facilitate nucleation or create conditions for slow crystal growth. This principle is the basis for the widely used industrial method of reducing grain size, known as modification. 2-2. Metal Ingot Structure The crystals which form in the process of solidification of a metal may have many different structures (dendritic, lamellar, needle-type or acicular, etc.) depending on the rate of cooling, and the type and amount of admixtures or impurities in the melt. Perfect crystals of proper external shape can be obtained only if crystallisation develops under conditions when the degree of supercooling is very slight and the metal has a very high purity. In the great majority of cases, branched or tree-like crystals are obtained, which are called dendrites (Fig. 20). The nucleus develops to form a dendritic crystal chiefly along the directions of maximum linear rate of growth. As a result, the long branches are formed first. They are the so-called axes of the first order (primary axes & of the dendrite) and they branch out in various directions from the initial nucleus. * The work An required to form a nucleus is equal to one third of the sum of the products of the area S of each face on the surface of the nucleus and the surface tension o on that iace.Metal Ingot Structure 33 While the primary dendrites grow in length, branches of the second order (m) evolve from their edges and grow in a perpendicular direction. Axes of the third order (n) evolve and on the second-order axes and so forth. A stage is eventually reached at which the spaces between the branches are filled with solidifying metal (Fig. 20). Upon further {freezing of the metal and development of the dendritic crystal all of the liquid metal in the spaces is solidified. Alter the complete solidification of the metal, dendritic crystals can sometimes be observed directly on the surface of an ingot, in a characteristic pattern, or on the surface of a pipe or shrinkage cavity, i.e., in regions with an insufficient supply of liquid metal. More often, dendritic structure is revealed only alter special etching of microsections (Fig. 21). Etching reveals not only the structure of the crystallites but the boundaries between them as A. well, Fig. 20. Dendritic growth (after D. K. The structure of a steel ingot is Chernov) shown schematically in Fig. 22. It is evident from Figs. 21 and 22, that three structural zones may be distinguished in the cast metal. Solidification of the liquid metal begins at the surface of the mould and proceeds, at first, chiefly in the thin layer of highly supercooled liquid adjacent to this surface. This leads to the formation of a zone of fine crystallites (due to the high degree of supercooling). A second zone of elongated columnar crystallites (zone of transcrystallisation) follows immediately the first zone. These stallites grow in the direction of heat removal, i.e., normal to the mould walls. The subsequent growth of the columnar crystallites from the mould walls proceeds by advancement of first-order branches into the molten metal and the evolution of higher-order branches as described previously (Fig. 20). In cases when the metal is highly overheated and then rapidly cooled, the zone of columnar crystallites may extend through the whole volume of the ingot {Fig. 23). Slow cooling of metal which is not overheated (as in castings of heavy cross-section) will create conditions favourable for crystal nucleation in the central part of the ingot. This will form a third structural zone in the central part of the ingot consisting of equiaxed tandomly-oriented crystallites (Figs. 21 and 22) of a size depending34 Solidification and Metal ingot Structure Fig. 21. Macrostructure of cast ° Fig, 22, Structure of a steel ingot (after -steel containing 0.8 per cent C N. A. Minkevich): s—plpe, 2—shtinkage cavities, $—shrinkage po- izath yeah Havent fe sqysteiies, bore columnar en, osama ‘celented ‘chystalltiee, /@7#*: Metal Ingot Structure 35 jon the degree af overheating of the metal, rate of cooling, presence admixtures, etc. The degree to which the various structural zones develop depends on the size-of the ingot, degree of overheating, rate oe ie Ot cohannas ‘erstallit high density (4 e zone of columnar allites possesses high density (few gas ‘cavities, blowholes, etc.}. The wtetal at the junctions of the columnar 'ystallites has a low strength, however, and in subsequent working dorging, rolling, etc.) fractures may occur at these places in the ingot. 23. Microstructure in the cross-section of a chromlum-nicke! stainless steel Pe mee “ingot (alter VM. Doroain) The development of columner crystallites is undesirable, for this rea- gon, in metals of low ductility, including steel. On the other haud, it is expedient to extend the zone of colurans: crystallites throughout the ingot for plastic metals, such as copper and its ailoys. High qucti- Hity ° an jlley excludes the fracture of the ingot in subsequent meeha- working. Due to the fact that liquid metal has a higher unit volume than solid meial and that solidification does rot take place simultaneously in the whole voiume, a cavity, called a pipe *, is formed in the upper of the ingot which freezes last: The pipe is usually surrounded the most contaminated portion of the metal. Alter solidification, 8 In rimming steel, the increase in volume due to the formstion of fas cavities es for ihe difference in the unit volumes of the iquid and colld :eetal and, fore, no shrinkage cavities are produced,% Solidification and Metal Ingot Structure micro- and macroporosity and blowholes are characteristic of this zone which is called the zone of shrinkage friability in the ingot. Gases evolve from the metal in the process of freezing since their solubility in the liquid metal is considerably higher than in the solid one. If cooling is sufficiently slow, surplus gases have time to leave the metal. They do not evolve at more rapid cooling and remain ia the ingot (or casting) forming so-called gas cavities or holes which may substantially decrease the mechanical properties of the casting or finished product.Chapter 3 PLASTIC DEFORMATION AND RECRYSTALLISATION IN METALS 3-f. Plastic Deformation Deformation is the change in dimensions or form of matter under the action of applied forces. Deformation is caused either by the mechanical action of external forces or by various physical and physico- chemical processes (for example, change in volume of separate crystallites in phase transformations or as a result of a temperature gradient). The process of deformation comprises the following consecutive stages: 1) elastic deformation, 2) plastic deformation, and 3) fracture. Elastic deformation is often defined as deformation which completely disappears as soon as the action of external forces ceases. Elastic deformation does not cause any noticeable changes in the structure of metals. The application of a load elastically displaces atoms to a slight extent, relative to each other, or twists blocks of the crystal. Tensile loads applied to a monocrystal tend to increase the interatomic spacing while compressive loads tend to reduce it. Due to the forces of tepulsion or attraction, the atoms return to their equilibrium positions as soon as the load is removed so that the initial form and size of the crystals are restored. It should be noted, however, that the elastic deformation of polycrystalline metals, especially at higher temperatures, causes so-called viscous flow and after the removal of the external forces, deformation floes not disappear completely even in the case of relatively small loads. In viscous flow, the grains composing the metal are not deformed; only certain crystallites are displaced in reference to others. At high temperatures and low stresses, viscous flow becomes much more important, but permanent changes in the dimensions and shape of a metal specimen occur only after long periods of stress application. This phenomenon is called creep. At stresses exceeding the elastic limit (the maximum stress that can be applied without producing a measurable permanent deformation or set after removing the stress) plastic deformation is observed. Plastic deformation is associated with displacements of the atoms within the grains and causes permanent changes in shape of the specimen. Plastic deformation in crystals occurs in two ways: by slip or by twinning.38 Plastic Deformation and Recrystallisation in Metals Slip is due to the action of shear stresses; block-like sections of crystals are laterally disptaced relative to each other on definite crystallographic planes which are typical for each type of crystal lattice. The critical resistance to shear ts less along these planes than the acting shear stresses. Numerous studies of the slip process have shown that it occurs firstly along planes and directions with the most closely packed atomic arrangement. In metals having a body-centred cubic lattice it Cr, W, Mo, etc.) the possible slip planes are the diagonal planes 101), in a face-centred cubic lattice (Cu, Al, Ni, Au, etc.) they wilt be the octahedral planes (111), while in a hexagonal system (Mg, Zn, tte), they will be the basal planes (0001). Slip planes and directions for different crystallic systems are shown in Fig. 24. The more possible slip planes and directions a metal has, the more plastic it is. 4 24. Ceystallographle possible slip planes (shaded) and slip directions (arrows) Slip planes and directions are subject to changes in accordance with the temperature, degree of preliminary deformation, and alloy composition. For example, in aluminium, subjected to deformation at low temperatures, slip will occur along planes (111); at high temperatures—along planes (100); Fig. 25 schematically illustrates the slip process in a monocrystal under a shear load. As a result of plastic deformation, the monocrystal is divided into layers or slip blocks which are displaced in reference to each other and are separated by thin fayers in which a considerable displacement of atoms has taken place. These intermediate layers with strongly distorted lattices are called slip planes. The slip process is not to be understood as a simultaneous displacement of the parts of a crystal. Such slip would require stresses considerably higher than those actually present in the process of plastic vrei propagates gradually along th section of a crystal and, agates gradually al e cross-section of a crystal and, at each mement, caly the toms ol a comparatively small portion of » the cross-section are being displaced. According to the latest views on plastic deformation, the sources of slip are groups of atoms, displaced relative to each other, i.e.,ay Plastic Deformation ES Fh: ‘dislocations. Slip originates at a definite point, where a dislocation exists, and proceeds by movement of the dislocation in the given plane ‘of the crystal due to shear stresses. The schematic diagram of slip deformation, given in Fig. 26, shows ‘the changes in the arrangement of the ators during the progress of islocation movement. At first only the upper part of the crystallite Fig. 25. Elastic and permanent deformation of a metal under the shear stresses: Seat crcaacd eluatic strain Foe acne Paces bs ane ron jing from toad above te, aa a permanent strain by 4 af Tissthe latte G—sbearing Wood removed. only pefuraneat steein remains "= Pig. 26. Progress of a dislocation scross a crystal .'is displaced to the left relative to the lower part. Plane ABCD, in ~~ this case, will: be an extra plane of atoms in the upper part of the crystallite. The edge AD of this plane, as mentioned earlier, is called ‘an edge dislocation (Fig: 26, a). :. 4 As is evident from Fig. 26, the edge dislocation is perpendicular to the direction of slip and should be regarded as the frontal boundary of that part of the slip plane along which local displacement as occurred. Fig. 26, > shows the changes that take place in the arrangement “of the atoms during the movement of the dislocation, Progress of «the dislocation through the whole crystallite to the opposite crystal lace causes a slip of one interatomic distance between the parts Mio.40 Plastic Deformation and Recrystallisation in Metals Dislocations may be either: 1) positive (|_)—in which the dislocation originates as a result of shear stress and the extra plane is in the upper part of the crystallite, or 2) negative (])—where the extra plane is in the lower part. In addition to edge dislocations, there are also screw dislocations which occur when the slip plane covers only a part and not the whole of the crystallite cross-section. A screw dislocation differs from the Fig. 27. Formation of a screw dislocation edge type in that it is parallel and not perpendicular to the shear vector. Fig. 27 schematically illustrates the formation of a screw dislocation (after W. T. Read). Let us assume that a part of the crystallite is displaced in the direction indicated by the arrow (Fig. 27). Disloca- tion AD appears as a result of incomplete slip. It is a screw dislocation, in this case, as is quite evident from Fig. 27, 6 where the arrangement of the atoms around the dislocation is shown *. Upon the formation of a screw dislocation, the crystal should be regarded as consisting of a single atom plane wound into a helical surface and not of parallel atom planes. A combination of edge and screw dislocations will give a composite dislocation. The mechanism for dislocation movement in the crystal, described above, explains why the process of deformation may proceed at comparatively low stresses at which simultaneous movement of all the atoms in a given slip plane would be impossibie. * The plane of the illustration is parallel to the slip plane.Plastic Deformation 4i Plastic deformation in such metals as gold, silver, nickel, cadmium, and certain others may also occur by twinning. Twinning is a plastic deformation which results in a change in orientation of one part of a crystal toa position symmetrical to the other part in reference to a separating plane called the twinning plane. As in slip, only very small parts of the crystal participate simultaneously in the twinning process. Plastic deformation of a polycrystalline metal proceeds similarly to deformation of a monocrystal. It must be noted, however, that in this case intercrystalline deformation is imposed on the process of plastic deformation inside the individual crystals. Intercrystalline deformation in polycrystalline metals is characterised by rotation or slip of the crystallites (grains). Plastic deformation of a polycrystalline metal is never uniform. This is due to the fact that the separate crystals deform in various degrees because of their dissimilar orientation. This random orientation of the grains affects the slip process in the metal as a whole. Upon the application of an external load, those grains deform first which have possible slip planes most favourably positioned in relation to the applied force. At the same time, the deformation of these crystals is impeded by their neighbouring crystals in which the possible slip planes are otherwise positioned. Therefore, the crystallites will be in a comp- lexly stressed state which hinders the development of slip throughout the whole volume. Considerable structural changes result from deformation in a polycrystalline metal. The first sign of permanent deformation in a metal is the appearance of slip bands or lines (Fig. 37). Slip bands are observed upon even the slightest deformations. As the degree’ of deformation is increased, the slip blocks in each separate crystallite turn to conform with the direction in which the force acts and the metal grains are elongated in the direction of deformation to form a so-called fibrous or banded structure (Fig. 28). Preferred crystallographic orientation of the grains, revealed by X-ray analysis, arises after considerable deformation. This regular crystallographic orientation of the crystals in relation to the external forces of deformation is called texture. The formation of a texture is accompanied by anisotropy of mechanical and physical properties. This is undesirable in most cases but is sometimes used to obtain optimum properties in a definite direction (see page 388). chem: Plastic deformation strongly affects al! physico-chemical properties of a metal. An increase in the degree of deformation *, as is evident from Fig. 29, increases those properties which characterise the resistance * The degree of deformation is determined by the formula: FoF 199 per cent, 0 in which Fy is the cross-section area of the test piece before deformation and F, is the same, but after deformation.42 Plastic Deformation and Recrystallisation in Metals Fig. 28. Effect of piastic deformation on the microstructure of » metal, X 300 {after S. S. Steinberg): ‘ a—structure belore deformation, 6—structure alter deformation of the metal to plastic deformation (ultimate strength o,, and hardness R,). The capability for plastic deformation (relative elongation 6, i per cent, and relative reduction’ of area wp, per cent), on the other hand, is reduced. This phenomenon_1s cal- led_strain or work hardening. Cold plastic deformation decreases the specific weight and the electrochemical potential but increases the chemical reactivity of a metal. Deformed metal possesses less corrosion resistance than undeformed metal. Cold deformation of iron substantially increases its coercive force and decreases its magnetic permeability. The residual inductiort is decreased with small degrees of deformation and increa- with high degrees, in the properties of metals subject to cold plastic defor- Fig. 29, Effect of cold working on mation (strain hardening) are asso- the mechanical properties of low-car- ciated with fragmentation of the ’ bon steel: mosaic blocks and an increase in the (hardness, 2—tenslle strength. $—elo- angles of their disorientation, the . appearance of considerable deformation stresses, and an increase in the number of dislocations. The smaller the blocks, the higher the strength of the metal will be. As a result of a high degree of deformation (about 80 per cent), the block will have a size from 200 te 400 A.Plastic Deformation 43 As mentioned earlier, a field of elastic stresses surrounds a ocation. Therefore, the external stress, required to produce slip, be sufficient to provide for movement of a dislocation counteracting field of stresses created by other dislocations or by stress fields at the boundaries of biocks, grains, particles of other 8, etc. ‘The dislocation density increases with the increase in the degree-of armation. This naturally leads to interference in dislocation ement and, consequently, increases the strength of the metal. eiore plastic deformation (in the as-annealed condition), the disloca- on density will be from 10* to 10° dislocations per cubic centimetre. P density in strain-hardened metal may reach 10!" per cubic centi- -[t may be noted here that a strength approaching the theoretical @-value may be obtained if no dislocations are present in the metal, e° In this case plastic deformation may be produced only by simultaneous displacement, relative to each other, of all the atoms located on sides of the slip plane. This condition has been proved experimentally by means of filament-like crystals called “whiskers”. The ultimate strength of a whisker of iron, one micron (0.001 mm) in diameter and 2.5 mm long, reached ac kg per sq mm and its elastic elongation was up to 5 per cent. Infortunately, only very small crystals of metal, free of dislocations, ean be obtained at the present time. However, extensive research in . this direction leads us to expect the production of whiskers of conside- .. Fable size in the near future. | . As mentioned above, any process of deformation will end in fracture if the stresses are increased to a high enough value. Two types of fracture are distinguished: brittle and ductile. * Brittie fracture consists in destroying the interatomic. bonds. by normal stresses. Brittle fracture is not accompanied by noticeable plastic deformation. The resistance to brittle fracture is called cohesive fh (S,). Brittle fracture is comparatively rare. For example, is type of fracture may be observed in zinc and its alloys or in and low-alloy steel at low temperatures, as well as when brittle layers : searegate on the grain boundaries. It must be stressed, however, that : pure brittle fracture is practically never encountered. Fracture is always ‘preceded by plastic deformation. Ductile fracture is due to t: tial stresses which have reached. a definite fracture value called the shear strength (t,). This type of fracture is preceded by considerable plastic deformation. Proceeding from the fact that S, and +, are constant values for a given material, Y. B. Friedman considers that brittle fracture, due * For a more detalled explanation see Morton C. Smith, Principles of Physical Metallurgy, Harper and Brothers, New York, 1956. -44 Plastic Deformation and Recrystallisation in Metals to normal stresses, will occur in cases where the cohesive strength is lower than the shear strength. If the tangential stresses produce considerable plastic deformation before the normal stresses reach the cohesive strength, the fracture will be of the ductile type. Thus, fracture due to normal stresses by separation or cleavage may be either brittle or ductile. Fracture due to tangential stresses, resulting from plastic deformation, will, of course, always be ductile. Most frequently, metal fracture does not occur as pure separation or shear but as a complex combination of these two types of fracture. 3-2. Recovery and Recrystallisation Plastic deformation, which distorts the crystal lattice and breaks up the blocks of initial equiaxed grains to produce a fibrous structure or {hin plates, increases the free energy level of a metal. Fig. 30. Effect of recrystallisation on the microstructure of @ metat (schematic), X 300 (after S.S. Steinberg): a—aftet deformation, b—after rectystallisation Deformed metal, in comparison with its undeformed state, is in a nonequilibrium, thermodynamically unstable state. Therefore, spontaneous processes occur in strain-hardened metal, even at room temperatures, which bring it into a more stable condition. The speed of these processes may be increased by raising the temperature. lf strain-hardened metal is heated to comparatively low temperatures, the elastic distortions of the crystal lattice are reduced due to the increase in amplitude of the thermal oscillation of the atoms. This heating will slightly lower the strength of the strain-hardened metal but the elastic limit and ductility will increase, though they will not reach the values possessed by the initial material (before strain hardening). No changes in microstructure are observed in this period. This oartial restoration of the original properties, produced byRecovery and _Recrystallisation 45 reducing the distortions of the crystal lattice without noticeable changes in microstructure, is called recovery. A further rise in temperature increases the mobility of the atoms. When a certain temperature (a definite value for each metal) is reached, equiaxed grains are produced (Fig. 30) and the metal is completely softened (Fig. 31). The formation of new equiaxed grains in the heating process, instead of the oriented fibrous structure of the deformed metal, is called recrystallisation. The temperature required for the beginning of 8 sa Ottimate strength 6,2 Ag/mm? Ratative etongation 3, % sage 2108 200 300 400 300 GOT WO 809 900 1900 Temperature, °C Fig. 31. Effect of temperature on the mechanical properties of strain-hardened copper recrystallisation is characteristic of each metal but depends on a number of factors and firstly, upon the degree of deformation. The higher the degree of deformation, the lower the recrystallisation temperature wi be According to the researches of A. A. Bochvar, the minimum recry- sta!lisation temperature of commercially pure metals, subject to high degrees of deformation, will be T,=04T, in which T, is the recrystallisation temperature and ‘» is the melting paint in the absolute temperature scale. When the minimum recrystallisation temperature (recrystallisation threshold) is reached, a sharp drop in the resistance to deformation (c,) is observed, in conjunction with an increase in ductility of the metal (5, per cent) (Fig. 31). The recry stallisation process usually proceeds in a certain temperature interval. First more favourable portions of the metal are recrystallised and then the process extends to the full volume. The presence of impurities impedes recrystallisation. The higher the temperature and degree of deformation, the more rapidly#6 Plastic D tion and Recrystallisation in Metals recrystallisation will . Recrystallisation, in essence, consists in having the atoms of the deformed metal overcome the bonds of the distorted lattice, the formation of nuclei of equiaxed grains, and subsequent t growth of these grains due to transfer of atoms from delormed to_und crystallites, . The nuclei of new recrystallised grains are most probably formed at places where highest distortions of the jattice are concentrated. As aresult of recrystallisation, the metal will consist of new equiaxed rains which are in a thermodynamically more stable condition. ie size of the recrystallised grains depends chiefly on the degree of mL NSS NSS i SS SS ~ eo wl PS " a a Fig: 32. Recrysteliisation diagrams: Stopper, b—-troa tous deformation of the metal and the temperature to which it is jeated. Relations between the size of recrystallised grains, degree of deformation, and the temperature are usually presented in the form of re- or isation diagrams (Fig. 32). The higher the degree of previous deformation and the lower the recrystallisation temperature, the smaller the grains will be. At low degrees of deformation (3 to 15 per cent), the grain size after recrystallisation may become very large (Fig. 32). This is called the critical degree of deformation. The formation of very large grains in this process sharply reduces the ductility of the metal. Therefore, the critical degre of deformation should be avoided. In the process of recrystallisation of such metals as iron, copper, and aluminium, the new crystals sometimes grow in an oriented affangement and the so-called recrystallisation fexture is obtained. This usually differs from the deformation texture but, in some cases, they may coincide (as in the recrystallisation of very pure aluminium). Recrystallisation. texture leads to anisotropy of the properties. ThisRecouery and Recrystaltisation a ers the fabrication properties of metals and, in particular, th: ability. Pig. 33, Macrostructure of deformed metal When a previously worked metal has compietely recrystallised, im the process of subsequent heating, certain recrystallised grains will grow at the expense of others due to migration of the grain boun-48 Plastic Dejormation and Recrystallisation in Metals daries. This process of growth of new recrystallised grains is called secondary recrystallisation. Secondary tecrystallisation is due to the tendency of the system to reduce its free energy. As mentioned above, the energy of atoms inside the grain is less than that of atoms on its surfaces. It is natural then that grain growth, which decreases the total surface area of the grains, will also reduce the number of atoms with higher energy and generally reduce the energy in the substance. Heat treatment, consisting in heating metal above the recrystallisation temperature to soften the strain-hardened metai, is called recrystallisation annealing. Deformation may be either cold or hot in accordance with its temperature as compared to the Tecrystailisation temperature. Cold deformation or cold working of a metal is a process accomplished at a temperature below that required for recrystallisation; cold working is accompanied by strain hardening. Hot deformation or hot working, on the other hand, takes place at a temperature above the recrystallisation temperature. In this case, hardening due to plastic deformation is completely eliminated by recovery and recrystallisation. This is true, however, only if the rate of recrystallisation is higher than the rate of deformation. An ingot is usually subject to hot plastic deformation. In the process of hot working, the dendritic structure of the ingot is destroyed and dendrites are elongated in the direction of deformation to form a fibrous structure (Fig. 33). The mechanical properties of a metal are anisotropic after hot plastic deformation. They are higher as a rule along the fibres than across them.Chapter 4 METHODS FOR STUDYING METAL STRUCTURE * 4-1. Macrostructure (Macrography) Macrostructure is the structure of metals and alloys as seen by the naked eye or by low-power magnification. Macrography reveals the size, form, and arrangement of crystallites (dendrites) in cast metals (Fig. 21), fibres in deformed metal (Fig. 33), shrinkage porosity and gas cavities (Table 1), cracks appearing during certain fabrication processes, chemical onthomogeneity in the distribution of certain constituents appearing in alloys upon their solidification from the liquid state (Fig. 34), etc. Macrostructure may be studied either directly on the surface of the work (a casting, for example) or on a fracture, or, more frequently, on specimens or samples cut out of large billets (ingots, forgings, etc.} or articles. The surface of the specimen is ground and then etched by special reagents. The action of etching reagents, or etchants, as they are called, is based on their ability to colour and dissolve various constituents in a different manner and to widen microcavities, cracks and similar defects. To reveal the dendritic structure of cast steel, for example, 2 10 or 20 per cent aqueous solution of ammonium persulfate [(NH,),5,0,} is rapa ed. Keshian reagent (25 parts H,04-10 parts H,SO,+ 65 parts HCl) is used to reveal fibres in deformed meta!, a5 well as cracks, cavities, porosity, and chemical nontiomogeneity. The Baumann metned is extensively used to reveal ihe degree of nonhomogeneity in the distribution of sulphur {or phosphorus) in steel. This method consists in applying ordinary silver bromide photographic Paper, soaked in a 5 per cent solution of sulphuric acid, to the surface of the sample. At the sulphur-rick areas, the peper will darken due to the formation of silves sulphide Ag,S. * Only a brief deswription of the methods used to reveal the structuie of metals ang alloys and the transiormations which occur in them is within the scope of this book. A mare detuiled description may be found In speciai Hierature, such as: G. I. Pogodin-Alexeyev, Y. A. Geller, and A. G. Rakhshtedi, Engineering Physical Metallurgy, Ovorongiz, Moscow, 1956.Principal Defects'In Cast and Deformed Steet (alter M. 1. Vinograd) Table £ Defect 1. Pipe 2, Central porosity 3. General porosity 4, Gas holes 5. Nonmetallic inclusions and internal] hair lines Metal has not filled the cavity in the uj upper Volume decrease (shrinkage) axial section of the ingot. The surface be during solidification in con- cavity is covered with oxides and adjacent regions are contaminated with inclusions down of furtl the head of insufficient feat reentlon by a shot lop" Fine pores, sometimes with sulphide or oxide] Formed by insufficient fi1- inclusions. Usually observed * on cles made of ling of" ts ental Part of zal or lower jo. ingot metal par ™ ‘solidification “ Inclusions or gas holes, located along the ee whole erony section of the article or at the end of the zone of columnar crystals. Have the form of dots, threadlike streaks, or hair cracks Channels or cavities not filled with metal.) Formed during Usually at the surface, at the end of the zone of columnar crystals, or randomly located over} gases (Hy, Ng, “and others) the cross-section of the article Presence of jmpurities in the metal, such as - particles of refractories, products of deoxidation of steel, etc. {alumina, silicates, sulphides, nitrides, and oxides). Certain nonmetallic poco ons of comparatively large size (larger tions) stretch in the direction of Serking ne ad produce internal hair tines Revealed by b Fracture hy 3. Xray examination 4. Ultrasonic flaw detec. ton 1. Fracture 2. Macrography J. Macrogra : 2. Magnet ec detection 1. Fracture ' : Stepped Tisning . Microgray . Mca en ne amas wiaWMicrostructure (Micrography) 51 A special case of macrography is the study of fractures in metals, This reveals the size and arr: ent of the grains, discontinuity flaws, etc. The character of the fracture obtained in cast aluminium ae ; ~ enter e EE a ctnemih fe *. Meee pepe en oe SS) Bara Fig. 34. Macrostructure of cast steel revealed by etching in a reagent composed of BS, FeCls, and CuCl, Nonuniform istribution of phosphorus and carbon is observed, alloys enables the amount of slag inclusions, coarse porosity, and other defects to be estimated as well. 4-2. Microstructure (Micrography) Micrography is the study of the structures of metals and their alloys under a microscope at magnifications from x75 to x 1,500. The observed structure is called the microstructure. The aims of metallographic study include the determination of the size and shape of the crystallites which constitute an alloy, revea!- ment of structures characteristic of certain types of mechanical working operations, discovery of microdefects (nonmetallic inclusions, micro- cracks, etc.), and, in some cases, the determination of the chemical content of alloys (for example, annealed carbon steels). Microstructure is an indication of the quality of heat treatment, of mechanical properties, etc. Micrographic_ investigations are conducted on specimens called microsections. These are prepared’ by cutting out a small specimen,a. Methods Jor Studying Metal Structure for example, a 10 mm cube of a washer 10 to 15 mm in diameter cand 15 mm high. One face of the specimen is smoothed with a file or emery wheel and then ground with special emery Paper arranged on thick plate glass or secured to a rapidly rotating wheel. The specimen .is ground first by the coarsest available emery paper and suc- cessively by finer and finer grades until the finest available grade ale Lg Fig. 35. Electropolishing Fig. 36. Microstructure of a sotid so- outit (shown schematically}: lution (Mg-Al) J—waler bath, 2—vessel_ with ele- ttrolyte, 3— microsection v cathode of sainiess Meet is reached. The direction of grinding Is periodically changed through 90°. After grinding, the specimen is polished by @ rotating disc, covered with broadcloth, felt, or velvet, to which a very finely ground abrasive material (alumina, chromium oxide, etc.) is continuously applied. : Electropolishing or electrolytic polishing (Fig. 35) is often employed for preparing microsections. Here the microsection 3 serves as the le and plate 4 of stainless steel is the cathode. A direct current with a density from 2 to 6 A per sq cm is passed through the electrolyte. Localised dissolving of the surface irregularities takes piace and a mirror surface is obtained on the microsection. Aqueous solutions of acids, mixed with solvents that facilitate dissolution of the anode products, are usually employed as electrolytes. The following electrolyte is most often used for electropolishing steel: orthophosphoric acid—83 millilitres, chromic anhydride— 11 grammes, sulphuric acid—15 millilitres, and water 5 millilitres. Good results are obtained for copper alloys with an electrolyte containing 980 grammes of orthophosphoric acid to 1 litre of water.4 Electropolishing reduces the time required to prepare a micro- Ketion and completely eliminates deformation in the surface layers Bye to grinding. Such deformation may distort the actual character the meta! structure. . se The finished microsection is first examined before etching to reveal pametallic inclusions, such as graphitic inclusions in cast iron. Microstructure is revealed by etching the microsection with a special nt called an etchant. Steel and cast iron are usually etched. with Fig. 37. Microstructure of iron: Slip bands are seen within the grains #4 to 5 per cent alcoholic solution of nitric or of picric acid. A 0.5 per ‘tent aqueous solution of hydrofluoric acid is extensively used for alu- ‘winium alloys and an 8 per cent ammonia solution of CuCl, is suitable lex copper alloys. Many other etchants are employed as well. Electro- ic etching has come into favour recently and is being widely used. is based on the different rates of electrolytic dissolution of the various & weuctural constituents of an alloy. e..” Ha pure metal or alloy, consisting of homogeneous grains, Is et- thed, the grain boundaries are revealed as a thin dark network (Figs. 36 f. -and 37). In some cases, grains of the-same composition are differently ‘tched because the different grains intersect the surface of the micro- fection along different crystallographic planes with varying chemical ‘activity. As a result, certain grains are almost unattacked and remain sight while others are heavily etched and appear darker, i.e., they ter light to a greater extent.BA Methods for Studying Metal Structure Deep etching reveals the so-called etch figures (Fig. 38) which indicate the crystallographic orientation of the grains. The various constituent structures of a nonhomogeneous metal are etched with varying degree of intensity. In the photograph of the microstructure of steel, shown in Fig. 39, the latter is composed in one case (a) of lamellae and in the other (6) of globules of Fe,C. By the action of the etchant on the microsection surface, iron (ferrite) is etched to a greater extent than iron carbide FesC (cementite) and a relief appears on the surface. Upon observation under a microscope with oblique illumination, the projecting boundaries of the iron carbide Fe,C produce a shadow effect (Fig. 40) and appear bright while the crystals of ferrite. on the boundaries with the carbides are dark. Metallurgical microscopes ate employed to investigate the structure of metals. Since the objects are opaque, they are observed in reftected light. a schematic diagram of the optical system of a metallurgical microscope is shown in Fig. 41. Optical microscopes enable only those details of the structure to be observed, that are larger than 0.15 to 0.2 microns (1,500 to 2,000 A).This corresponds to a useful magnification from Xx 1,000 to x 1,500. At the present time, the electron microscope * (Fig. 42) is also used to study metals and alloys. An effective magnification of x 100,000 may be obtained (magnifications from x 7,000 to x 25,000 are usually employed). The use of an electron beam with very short wave length (0.04 to 0.12 10-* cm) enables much smaller particles to be observed than in optical microscopes. Details of the object may be distinguished which are from 30 to 50 A in size. This is about 100 imes smailer than can be seen in an optical mnicroscope.Electromagnetic or electrostatic fields are used as “lenses” to focus the electron beam. The source of the electron beam is the so-called electron gun consisting of a tungsten filament which emits electrons when heated to a high temperature. The electrons are accelerated in a strong electric field and are directed to the object. The actual surface of the metal is not usually observed in the electron microscope. A quartz or collodion replica, made from the etched surface, reproduces all details in its relief that depend on the actual structure of the metal. The electron microscope has considerably extended the field of application of micrography. It enables the structure to be determined in cases when the high dispersity of the phases present in the metal makes observation through an optical microscope useless. Vacuum metallography is being employed in the last years. It in- volves the study of microsections heated in a vacuum. * The electron microscope and methods of electron microscopy are described in detail in A. I. Gardin, Electron Microscopy of Steel, Metailurgizdat, Moscow, 1954.Microstructure (Micrography) MOR | Fig. 39, Microstructure of steei containing 0.8 per cent C, x 400: a—FecC precipitated as lamellae (lamellar pearlite), b—as grains (globular peartite). a Pig. 40. Diagrams explaining features in the structure of pearlite as seen under 3 microscope: 4—lamellar peariite, ’—granuler pearlita (aiter F. 1. Masleanikov) f 2 4 le. tT wn 3 1 j—— a / 5 Pig. 42. Diagram | of Fig. 41. Optical da electron microscope: |__patiges! masopes *f faite afta? Sh ‘intron, “Epi, ans aac cl Bea,X-Ray Analysis 57 itraviolet microscopy, developed in the U.S.S.R., is successfully d to investigate the structure of complex multiple-phase alloys. ‘this case, in place of visual observation, mictophotography is plied in ultraviolet rays* . The latter. are invisible to juman and have a wave length about 1.5 times shorter than visible At. For ultraviolet microscopy, either unetched microsections ge used ot microsections coated with oxide films. A special ultraviolet Rroscope is used. It has a quartz glass optical system, transparent to jolet light. The source of light is a mercury-vay lamp which ively produces ultraviolet light. The principle of ultraviolet scopy is based on the fact that the reflection factor of the diffe- phases and regions of the metal varies for the ultraviolet rays fall- feg.on the specimen. : pMlagnetic microstructural analysis has been introduced recently, is based on the Principle that ferromagnetic structural constituents loys attract the particles of a magnetic colloid while nonmagnetic stituents do not. Therefore, the distribution of ferromagnetic con- Mituents in an alloy. may be observed under a microscope by the di- fribution of the magnetic particles which form a so-called pattern. = 4-3. X-Ray Analysis -ray analysis is extensively employed to study the crystalline ture of metals (X-ray crystallography) and to-teveal internal de- (X-ray flaw detection). “rays ** are a form of elect: etic. radiation but have very wave lengths. Their wave enh is of the same order as the watice constants of metals (0.1 to 10 A). The most interesting feature i their ability to penetrate materials opaque to visible light. X-rays e described as “soft” if they have a larger wave length and are more lly. absorbed by materials and as “hard” if they have short wave gths and a high penetration. X-rays are usually obtained by slowing a rapid stream of electrons on the surface of the anti-cathode in ial X-ray tube. X-ray tube is a high-vacutim glass bulb into which two elec- les have been fused. One is the anode (or, more properly, anti- eatho de} and the other is the cathode. The source of electron emission Pie a heated tungsten filament. The electrons, due to the high potential Péhaintained between the anti-cathode and the cathode, at a high ‘yelocity to the former and interact with the atoms of its material. is interaction excites the electrons in the atoms of the anti-cathode lerfal. Conversion of these electrons back to a lower energy level ates the electromagnetic radiation called X-rays. Tubes for X-ray * Photography is also done in visible rays to obtain a white image. - Discovered in 1895 by W. Roentgen. " ay58 Methods for Studying Metal Structure flaw detection operate at voltages from 200 to 300 kV (X-ray tubes for 2,000 kV are also made). Tubes for structural analysis operate at voltages up to 50 kV. A tungsten anti-cathode is used in tubes for radiography having arating up to 3 kW. Tubes for structural analysis, rated up tol kW, have anti-cathodes of chromium, iron, cobalt, nickel, and other metals. X-ray crystallography is widely employed to ascertain the relative arrangement of atoms in metals, to determine the Phase composition of alloys, to study recovery 2 and recrystallisation processes, ¢ to define internal stresses (dis- P. ' tortion) in the lattice, grain 4 size, etc. This method of analysis is based on an interference 5 if a pattern of rays diffracted by each atom of the material se- Fig. 43. Diagram illustrating the Wolfe. parately. This complex pheno- Bragg equation: P-P’ and QO’ are atomic menon of the interference of planes X-rays, diffracted by the atoms of a crystalline solid, procee:'s in the same manner as if the rays were refiected from parallel atom planes of the crystal. The condition for the production of a reflected X-ray beam is specified by the Wolfe-Bragg equation (Fig. 43): 2d x siné = nd in which d is the distance between the parallel atom planes (inter planar s acing): ® is the angle of incidence of the beam on the atom plane; 4 is the wave length of the X-rays; a is an integral number (so-called reflection order). The Debye and Scherrer method of structural analysis of polycrystalline samples is most often employed in physical metallurgy. It is also called the powder method. The principle involved to obtain a structural radiograph of a polycrystalline specimen consists in having a fine pencil monochromatic X-rays strike the specimen P (Fig. 44, a). In the crystals of the specimen, which are randomly oriented in reference to the X-ray pencil, there are always a considerable number of atomic planes of such orientation as to reflect the X-ray pencil, since the angle 6,'at which the beam strikes them, conforms to the Wolfe-Bragg equation. At a definite angle 6, the beam reflected from any of the crystals will evidently pass along the elements of a cone whose axis is a pro- longation of the pencil. The apex angle of the cone wiil equal 4 @ (Fig. 44, a). Reftections of the first order will be obtained for planes for which th. Planes with a larger interplanar spacing d produce several cones of reflected radiation of varying order.X-Ray Analysis 59 If the number of reflecting crystals is very large (practically, when their size does not exceed 1 to 3 microns) the cones will be continuous. A flat photoplate film set at a certain distance behind the specimen, perpendicular to the direction of the X-ray pencil, will show a series of concentric rings which represent intersections of the cones by a plane. Usually radiographs of polycrystalline samples are obtained not on a photoplate, but on a compa- Tatively narrow strip in a cy- lindrical holder surrounding the specimen. The radiograph, 0 . then, consists of a series of Photographic arcs (Fig. 44, 6), symmetrical Sela in pairs, ‘ To describe the crystal lattice it is necessary to determine the size and form of the unit cell, the number of atoms it contains, and the coordinates of each atom. For this purpose, the angles of reflection of the planes and their corresponding indices are determined by the distance between the symmetrical lines on the fadiograph. The intensities of the Debye lines are also determined *. A detailed analysis ofa radi: ogtaph Is not often required in pi) 44. Method of obtaining a radi practice. Sometimes it will be eon of a polycrystalline specimen: Fence ice Mattern ae Sut ge Maced eatin, one ical diffraction pattern of the He airbye-Scherre metal being investigated and "*S°# 2?" (rasten Rater) obtained by tnis to compare it with the radiograph obtained in a Debye camera. Each ring (pair of arcs) on the radiograph is a result of diffraction from planes with definite indices (Ak!) in different, suitably oriented, crystals. If the structure of a certain crystalline substance is known, the positions of the rings (lines) on the radiograph may be computed, as well as their relative intensity. * Methods used in measuring and analysing radiographs are given in special literature; see, for example, Y. S. Umansky, A. K. Trapezriikov, and A. I. Kitaigo- rodsky, Radiography, Mashgiz, Moscow, 1951.CJ Methods for Studying Metal Structure To determine the content of various phases in an alloy, it is sufficient to make a radiograph and to determine whether’ it has s: j- cal ines spaced at a distance equal to that computed beforehand for the most intensive lines of the phase being determined. The number of lines on a radiogra , their form, position, and uniformity of dar- kening enable conclusions to be made concerning the changes that have taken place in a metal or alloy. Thus, the appearance of-a new phase will produce additional lines on the radiograph, an increase in grain size will cause a lack of continuity or gaps in the lines, formation of a texture will darken the Debye lines to the maximum extent, etc. In addition to. X-ray crystallography, electronographic analysis is also widely used in laboratory practice. The _pigiomence of diffraction is produced by.the passage of a pencil of rons thre acry- stalline solid ia the same manner as for a pencil of X-rays. Electron diffraction can be expediently applied as a method of investigation in the same cases that X-ray structural analysis is used. Metal structure is revealed by electron diffraction in a special instrument called an electronograph. A narrow pencil of accelerated electrons, obtained in the electronograph, is directed to the metal specimen and the resulting diffraction pattern is recorded on a Photographic plate or on §lm. Electronographs of latest designs enable diffraction patterns to be observed at different temperatures. This ides for a direct study of phase transformations in metals and al cys. In principle, the electronograph resembles the electron mi Fi . 42) which is frequently used to obtain electron diffraction patterns. : RADIOGRAPHY (X-RAY FLAW DETECTION) Radiography is extensively employed to inspect cast, welded, or forged components for the purpose of detecting internal flaws (blowholes, porosity, lack of weld penetration, etc.). A schematic diagram of an X-ray flaw detection unit is shown in Fig. 45. X-rays, passing through a solid, are partially absorbed by the material. The absorption will vary along the cross-section, if internal defects are it, and therefore a photographic film, plsced so as to intercept the emergent beam, will show lines or areas of varying intensity. The negative will be darkened more at places corresponding to the ibcation the defects, due to less absorption of radiation.* Up-to-date X-ray units provide radiation with a penetrating power that enables steel specimens up to 100 mm thick, copper and its alloys up to 60 mm thick, and aluminium and its alloys * Regions more penetrable to X-ray radiation will be darker, areas of high absorption ill be lgtter ne (OXY ’ ‘ehMethods jor Studying and Inspection of Metals ond Alloys 6t to 400 mm to be tested. The y-rays of radioactive isotopes ye better results for thicker sections, The atom nuclei of such are of unstable structure and spontaneously emit much in the form of a-particles (helium nuclei), f-particles lectrons), and y-Tays. These y-rays resemble X-rays but have el 8 te Ua? Fig. 45. X-ray flaw detection unit: tube, 2—direetion of beam, 3--specimen, 4—casselle, 5—X-ray film, 6—lead shield, 1G Jiaclusion tol higher denalfyi, Bec huorescent xeon, 702 ray. Blas or teen i. en shorte? wave lengths (from 10~** to 10-"! cm).. This makes them Emore penetrative than X-rays. The source of y-radiation is usually bathe radioactive isotope of cobalt, enclosed in a special ampule. 4-4. Physical Methods for Studying and Inspection of Metals and Alloys &. Thermal analysis.* Phase transformations in metals and alloys ate accompanied either by the evolution or absorption of heat. Ther- mal analysis consists in determining the temperature at which heat is absorbed or evolved. This can be done. by simply measuring the ~_——_—_ * The development of thermal analysis has been based, to a great extent, on the work of Academician N.S. Kurnakov, the outstanding Soviet chemist.62 Methods for Studying Metal Structure temperature of the metal (or alloy), being heated or cooled, at equal intervals of time. This temperature-time relationship is plotted as a heating or cooling curve (Figs. 70 and 81). Broken lines or horizontal steps on the otherwise smooth curves indicate the temperatures at which heat is evolved (in cooling) or absorbed (in heating) due to transformations in the alloy. Temperatures at which changes occur in an alloy are called the critical points. Temperature difference between specimen and standard S $1000 ie ye S 400) i : KR Time Fig. 46. Heating and cooling curves tor a metal: Curve / is plotted in “temperature- time" coordinates while curve 2is plotted in “temperature difference (between test- piece and standard)-time” coordinates This type of thermal analysis is applicable only if a considerable amount of heat is evolved. Differential thermal analysis is employed to record transformations accompanied by very small thermal effects. In this procedure, the difference in temperature is determined between the specimen being investigated and another material (standard), which does not undergo transformation in the range considered. Both are heated or cooled simultaneously, under identical conditions. The results of a differential thermal analysis are plotted as temperature difference-time curves (Fig. 46). At the moment of transformation,Physical Methads for Studying and Inspection of Metals and Alloys 63 the temperature difference increases sharply and does not depend on the healing or cooling conditions. Dilatometric analysis * is also used to determine the temperature of phase transformations (critical points). For this purpose, changes in length are noted in heating or cooling the specimen, or under isothermal conditions (constant temperature). Changes in the length of the specimen indicate the vo- Jumetric changes that accom- any a transformation in the metal (Fig. 47, a). This method is applied for determining the critical points in steel since its transformations are always accompanied by substantial changes in volume (due to the different specific volumes of a- and y-iron). The differential method of dilatometric analysis is used to obtain higher accuracy. In this case, the difference in the expansion between the specimen and the standard is measured in reference to the temperatu- te. The temperature is determined by a dilatometric pyrome- ter, ie., by the expansion of astandard material having a known relation of expansion to temperature. A dilatometric curve for an 0.8 per cent _C steel is illu. flrated in Fig. aT, When the temperature reaches a cri- ., . tical point in heating a speci- Fig. 47. Oates ees et steel con- men, the changes in its length a—simple record _in “extension-lemperature” co- as differing from the standard Sigimlet (i diernliat eed op, agrsion will depend, not somuch upon perature” coordinates thermal expansion (in heating) or contraction (in cooling), as upon the transformations in the metal or alloy (for example, allotropic transformations). At this point an abrupt expansion or contraction of the specimen occurs. This produces a sharp break in the hitherto smooth differential curve which turns sharply upward or downward. o Length emperature Sta Po al. ai, * Dilato is the Latin for “1 expand”.ry Methods for Studying Metal Structure Special instruments, called dilatometers, are used in dilatometric analysis, They mfy also be employed to determine the coefficient of thermal expansion in a given temperature interval. Electrical research methods, The physical parameter determining the electrical properties of metals is their specific resistance or electrical resistivity p, expressed in ohm-mm?/m. The electrical resistivity of metals and alloys not subject to phase transformations in heating, increases upon an increase in temperature according to the linear equation: P= Pn [lta (t,—t)) in which Is the temperature coefficient of resistivity. The resistivity of metals and. alloys depends on their specific physical nature and may chi substantially in accordance with their composition and structure (this was first discovered by N. S. Kurnakov). The resistivity of alloys differs considerably from that, of pure metals (Figs. 85, 89, and 94). Its variations, depending on the concentration of the elements comprising the alloy, are essential indications of its nature and structure. Electrical resistivity of alloys is ordinarily measured by the duplex Thomson bridge or by potentiometric methods (which are more ac- urate). Magnetic methods of research and inspection. The chief characteristics of the magnetic pre jes of materials are the permeability p, which is defined as the ratio of the induction B (in gs) to the magnetising force A (in oersteds): a b= and the magnetic susceptibility K K=4 in which / is the intensity of magnetisation. The Suseeptibilit K Is related to the lity » by the equation h=1+-4aK. y Depending on the character of their magnetic properties, materials are classified as: . 1. Diamagnetic materials, having a small negative susceptibility, independent of the temperature and magnetising force. Such materials include zizic, gold, nereury, ete - 2. Paramagnetic materiats, for which K>0. These include lithium, sodium, rubidium, and other metals, 3. Ferromagnetic materials, such as ion, nickel, cobalt, and their alloys. These metals have a positive magnetic susceptibility which greatly exceeds that of paramagnetic materials. The permeability of ferromagnetic materials may reach high values {up to 3.5x 10*) and depends not only on the temperature but also upon the magnetising forcePhysical Methods Jor Studying and Inspection of Metals and Alloys 65 etic materials are characterised by the initial permeability the maximum permeability p..,- The relationship between the of magnetisation and the magnet ising force or field is illustrated magnetisation curves and hysteresis loops (Fig. 48). . magnetic induction 8 first increases with an increase of the ising feree H and then remains practically constant. Fig. 48. Magnetisation curve .and hysterests loops “The phenomenon of hysteresis is typical of ferromagnetic materials. When the magnetising force is reduced to zero, a ferromagnetic mate- eh magnetised to saturation, will retain residual induction B, (Fig. 48). : The reverse magnetising force required to reduce B, to zero is called F the coercive force H. (Fig. 48). The magnetisation of ferromagnetic materials decreases as the ; temperature increases. Above a definite temperature for each given : metal (Curie temperature), ferromagnetic properties disappear and :. the metal becomes paramagnetic. When a ferromagnetic material is placed in a magnetic field, a small change in linear dimensions and volume results. This phenomenon Is called magnetostriction. The saturation intensities of magnetisation and magnetostriction are - structure-insensitive properties. They are determ by the composi- ‘tion, atomic structure, and the quantitative ratios of the phases of ‘which the alloy is comprised. Coercive force (H,), permeability (1) or susceptibility (K), and residual induction @) are structural- sensitive ies. These ‘ies depend to a great extent upon > the dispersity and shape of the crystallites, their orientation and relative arrangement, as well as distortion of the crystal lattice, i.e., upon stresses, etc. (see Sec. 13-12). ~ A study of the changes that occur in magnetic properties when famposition and treatment are varied enables one to understand the transformations taking place in ferromagnetic metals and their66 Methods for Studying Metal Structure alloys. Magnetic research methods are extensively employed, for example, in studying allotropic transformations in iron and its alloys. They are particu arly useful to study the transformations of supercooled austenite (Sec. 8-3) and for determining the amount of retained austenite (Sec, 8-5). They are also used to study the ordering of solid solutions (page 108), the ageing process (Sec. 8-7), and others. The magnetic method is also used for nondestructive testing of the composition and structure and, therefore, the properties of many articles after heat treatment. The testing methods are based on the measurement of a definite magnetic characteristic such as: magnetic saturation, residual induction, coercive force, and magnetic permea- ility. A magnetic testing method wil! satisfy all requirements of shop inspection for those cases in which a definite. relationship has been established between the magnetic properties and the structure of steels. Magnetic testing is distinguished for its high sensitivity, rapidity, simplicity, and the possibility of being fully automated. lagnetic inspection (magnetic flaw detection) is most extensively used to reveal surface defects on machine components (quenching, grinding, and fatigue cracks, hair cracks, lack of weld penetration, etc.), The method is based on the principle that when a ferromagnetic component is magnetised, stray fields are produced at all discontinuities in the surface (at cracks). If the component is immersed in an oil (or gasoline, alcohol, etc.) bath containing suspended particles of magnetic powder (magnetic suspension), the particles will be at- tracted by the dispersion field and will collect on the edges of the aw so as to reveal jit. The radioactive isotope method (tadioactive tracers) has been widely used in physical metallurgy in recent years. The application of radioactive tracers (“tagged” atoms) became ible after the discovery of artificial radioactivity by Frédérick Joliot-Curie and [rene Curie in 1934. The atomic nucleus of a radioactive material has a complex and unstable structure. Therefore, such materials continuously and spontaneously emit considerable energy in the form of radiation that cannot be controlled. As mentioned above, this radiation includes: 1) a-rays which are positively charged helium nuclei and are only slightly deflected by a magnetic field. When a-rays pass through a gas, it is ionised. This enables the passage of the rays to be observed by their traces or, as. they are called, “tracks”. The velocity of «-particles is about 17,000 km per hr. 2) B-rays which are negatively charged electrons and are strongly defected by magnetic fields. B-rays possess Jess ionising capacity than a-radiation. The velocity of f-particles approaches that of light. 3) y-rays which are electromagnetic waves of the X-ray type but with considerably shorter wave length. y-raysPhysical Methods for St and Inspection of Metals and Atloys 67 are undeflected by a magnetic field, they are not charged, and do not jonise gases. Radioactive or unstable isotopes are the ordinary source of radiation. Radioactive isotopes of most of the elements may be obtained artificially as by-products of nuciear reactions. More than 1, fadioactive isotopes have been obtained artificl- ally at the present time. Radioactivity is associated with processes which take place in the atomic nucleus. The number of atoms that decays in a unit of time is proportional to the initial number of radioactive atoms. The same fraction of the radioactive atoms, strictly defined for each element, decays in each time unii. This fraction is known as the radioactive decay constant. The decay constant determines the possibility for decay or disintegration in a unit of time. The larger this constant, the more unstable the atomic nucieus is. Another value is usually employed to characterise the rate of radioactive decay. It is called the half-life period and is equal to the time tequired for the initial number of radioactive atorns to be reduced by one half. The half-life of various elements ranges fromm 10-° second to several milliard years. The half-lives of sertain more frequently used isotopes are: Fe'*—57 days, S*#—&8 days, C'—from 10° to 105 years, B!7— 0.022 second, etc. The application of tagged atoms or tracers for studying metals and alloys is based on mixing radioactive and nonradioactive substances. The introduced radioactive substance cannot be chemicaily separated from the nonradioactive substance and they act in exactly the same way in all processes that occur in the metal. Due to their tadioactivity, however, movements of these tracers can be observed. Special counters are used to register radioactivity (radioactive radiation). They are based on the ionisation uf the substance in the detector or tube of the counter by radioactive particles. The counter enables each elementary act of decay of a rudivactive substance to be registered. The autoradiography method may also be used for the same purpose. Here, a specimen containing the radioactive clement is ET on a special photographic plate (rhetographic emulsion). he location of the blackened areas, that appear fier the plate is developed, corresponds to the distribution of radioactive substance in the specimen. The degree of blackening or density is proportional to the amount of the isotope in the given volume. These blackened areas or spots are due to the action of radioactive emission on the photographic emulsion. The photograph obtained by this methad is called a micro-autoradiogramme (Fig. 49). Due to its high sensitivity and accuracy, the radioactive tracer method is finding more and more extensive application. In physical metallurgy, this method is successfully used for studying the ptocess of diffusion (self diffusion), the distribution of impurities and of elements speclally introduced into an aiioy, and also for phase 5t68 Methods jor Studying Metal Structure analysis. Radioactive tracers are of exceptional value for studying the mechanism of friction and wear in metals *, as well as the behaviour of metals in forging and press-working.** Ultrasonic inspection *** is being employed to a certain extent, lately, for revealing al! sorts of internal fiaws (blowholes, porosity, certain cracks, etc.). Fig. 49. Microstructure (a) and micro-autoradiogramme (b) of the same specimen Ultrasonics are elastic waves of frequencies far beyond the audio range. Frequencies of the order of 10* cycles per second are used in ultrasonic flaw detection. . The jditectivity of Ate vibrations werent pith an increase in their uency. a ler of one megacyele ‘second, the je of di noe 1s 30 small that th may be called Pittrmonic beams Co omy Ultrasonic inspection is based on the property of ultrasonic vibra- * BLP. N |. Rewarch on Wear in Culting Toots Conducted with Redtoactice Isotopes, 7 1965, uss.R. vee evga Pracesdings of the Academy of Sciences of the “sas F é a more Getall see Up-toDate Methods of Testing Exginsring Materials,wee Physical Methods for Studying and Inspection of Metals and Alloys 6 tions for passing through metals over large distances, as directed from the surface of various defects. A flaw, deep within the metal, may be detected and its shape may be determined by registering the reflected waves or the energy of the waves that have passed thi the metal. f General methods employed in ultrasonic flaw detection are the ollowing: 1. The pulse method. This more extensively used method is designed | for detecting discontinuities in the metal (blowholes, porosity, etc.), located at depths from several millimetres to several metres, and. for establishing coarse-grained zones in blanks and finished articles: The reflection of ultrasonic waves from the boundaries of flaws is used to reveal blowholes, porosity, and cracks. Coarse-grained zones are detected by the scattering of the waves by large crystallites. In the operation of a pulse-type ultrasonic flaw detector, the emitter * transmits short pulses of elastic vibrations (with a duration from 0.5 to 10 microseconds) separated b pauses from 1 to 5 milll- séconds tong. Upon the presence of an ol {flaw in this case), which may be considered as a dlisconf inulty in the acoustic properties of the medium, the echo-signal is reflected from the boundary of the flaw to the receiver and is registered by an indicating device. The distance to the defect may be determined by.the time that elapses between the transmission of the * pulse and the reception of the echo- signal. If the zone being inspected has a coarse-grained structure, the ultrasonic vibrations will be ‘scattered at the boundaries of the crystals. The echo-signal will be considerably weakened and may even disappear completely. The pulse method is used when the inspected item is accessible from one side only. 2. The resonance method is based on the generation of a standing wave in the metal due to interference. In this method, the variations in the operation of the emitter are observed at the moment that standing waves appear. The resonance method is chiefly employed for testing items up to 15 mm thick. It is also used to measure the thickness of metal, zones of corrosion, etc. 3. The shadow method is based on the “sound shadow” which appears beyond a defect located in the path of ultrasonic vibrations. This method has found comparatively few applications due to the cumber- some apparatus required and to a number of inconveniences in ope- tation. This method, however, may be further developed to obtain _ 4n image of the defect on a screen. In conclusion, it must be noted that ultrasonics are being employed more and more extensively in the Processing of metals to speed up diffusion processes, to change the conditions of solidification (and consequently, the structure of metals) in the casting of metals, for cleaning - metal items, etc. cA piezo-electric ‘Wansducer excited by radio irapulses of a special generator.Chapter 5 THE MECHANICAL PROPERTIES OF METALS 5-1. Mechanical Testing of Metals In the course of operation or use, all articles and structures are subject to the action of external forces which create stresses that ine- vitably cause deformation. To keep these stresses, and, consequently, deformations within permissible limits, it is necessary to select suitable materials for the components of various designs and to apply the most effective heat treatment. A comprehensive knowledge the chief characteristics of semi-finished metal products and finished metal articles (strength, ductility, toughness, etc.) is essential for this purpose. For this reason, the specifications of metals, used in the manufacture of various products and structures, are based on the results of mechanicai tests. Mechanical tests are those in which specially prepared specimens (test pieces) of standard form and size (and less frequently, the articles and structures) are tested on special machines to obtain the strength, ductility, and toughness characteristics of the metal. Mechanical tests may be conducted under various loading conditions. Tests may be: 1) static, when the load is increased slowly and gradually and the metal is loaded by tension, compression, torsion, or ending: 2) dyna- mic, wher the load increases rapidly as in an impact, and 3) repeated or fatigue (both static and impact type), when the load repeatedly varies in the course of the test either in value or both in value and in direction. 5-2. Tension Tests Tension tests are of the static type, i.e., they are tests in which the load is increased comparatively slowly, from zero to a certain final value. Standard specimens (GOST 2055-43 *), used for tension tests, are shown in Fig. 50. The specimens may be of round (a) or flat (6) cross-section. The ends of the specimen are secured in the grips of the testing machine. The mechanical properties in tension are determined on the gauge length /, of the specimen. Such testing machines as the universal lever-pendulum type (Model P5), the Gaga- * Here, and throughout the book, GOST (Russian—TOCT) fs the abbreviation for the U-S.S.R. State Standard,Tension Tests 7m rin press, model HM-4P (designed by the TsNITMASh Institute), and others * are extensively used for tension tests. All tension testing machines have two chief parts: 1. unit for applying a load to the specimen (with a hydraulic or mechanical drive), and 2. unit for measuring the applied load (lever with adjustable weight, pendulum-type lever, pressure gauge and hydraulic capsule). Testing machines are often equipped with recording devices which Fig. 50. Tensicn-test specimens. The gauge length fy for_a long specimen is taken as f=11.3 V Fo and for a short specimen f5=5.65 Y Fy, in which Fo is the cross- sectional area in sq mm. The standard dimensions of the long specimen should be: f5=200 mm, d=20 mm, Fy=3i4 mm? automatically record changes in length of the specimen in accordance with the applied load (stress-strain diagram). The most accurate tension stress-strain diagrams may be obtained on the Gagarin press and the model HM-4P testing machine. Fig. 51, @ illustrates a tension load-elongation diagram for low- carbon steel. It is evident from the diagram that at first the elongation is proportional to the load. This law of proportionality remains valid up toa certain load P, which is called the load at the limit of propor- ionality. The stress, calculated on the basis of the initial cross-sectional area of the specimen P Opa kg per sq mm is called the timit of proportionality or proportional limit. The conventional proportional limit o, is the stress at which the deviation from a linear relationship between the stress and deformation has reached a value for which the tangent of the angle between the curve and the stress axis has increased by 10 per cent. Stresses not exceeding the proportional limit result practically in elastic deformation only, i.e., strain that disappears when the load is removed. For this reason the proportional limit is frequently identified with the elastic limit, so that o,=o,. This is actually not so but accurate enough for practical putposes. Usually the elastic * A detailed description of these and other machines is given in D. O. Slavin and D. M. Shapiro, “Mechanical Testing of Metals, Metallurgizdat, Moscow, 1950.T2 The Mechanical Properties of Metals limit 9, is defined as the stress at which the permanent. set is equal to 0.005 per cent of the initial gauge length i +) of the specimen. Thus, o, and 6, characterise the resistance of the metal to. small plastic ‘deformations. The, proportional and elastic limits are essential characteristics of a metal. Any structure or machine component must be designed so that its working stresses do not exceed 9, and 6,. an 5, g/mm® 44am Strata, & Fig. 51, Tension-test diagrams There is a line relation in the region of elastic deformation between es:.. id strain for metals and alloys. [t conforms to the law of pro- ty (Hooke's Law): The coefficient of proportionality E called the modutus of elasticity (Young’s modulus) characterises the rigidity of a material, i.e., its resistance to elastic deformation in tension. The modulus of normal elasticity E depends only to a comparatively smal! extent on the structure, heat treatment, and composition of an alloy. It is determined, mainly, by the type of ‘crystal lattice. The moduli of elasticity for certain important engineering metals are as follows: : 7,200 | 12,100 | 21,000 | 4,500 | 10,300 | 36,000 Tron, which has a high modulus of elasticity, is the most important of engineering materials. - At an increase of load beyond P,, the relation between load and elongation will depart from a straight line. The straight line will. Tension Tests 73 become a curve and, at a certain load, a horizontal step will some- ’ times appear on the diagram. This horizontal step indicates that the metal is elongated (yields) without an increase in load. The load P, is called the load at the yield point and the corresponding stress -?. 2 a, i kg/mm! * is called the yield point. Thus, the yield point is the minimum stress at which the specimen is deformed (yields) without a noticeable increase in the load. Appreciable plastic deformation, i.e., deformation which remains . after the load is removed, occurs above the proportional limit. Elastic deformation causes only elastic distortion of the crystal lattice while in plastic deformation, slip of some magnitude takes place. : metals do not have a clear-cut yield step (“sharp-kneed” } stress-strain diagram).* In this case, the curve passes smoothly from the elastic section to the section corresponding to plastic deformation (Fig. 51, 6). For such diagrams the conventional yield point is defined as the stress at which the specimen receives a permanent elongation equal to 0.2 per cent of the initial gauge length denoted as ,.,. The . yield point is easily determined and, for this reason, is extensively employed as an indication of strength in engineering materials. At the present time, it has often supplanted such characteristics as o, and o,, since the latter require precise instruments for their deter- ‘mination. A further increase in load will cause more marked deformation in the whole volume of the metal. The maximum load which the specimen . can withstand without failure is called the load at the ultimate strength and corresponds to the stress o,= e kg/mm? which is called the witimate strength or tensile strength. » The ultimate strength o, is the stress corresponding to the maximum load before failure of the specimen. Beginning with the stress » G,, the deformation is concentrated in one section of ductile metals, A localised reduction of cross-sectional area or “necking down” appears on the specimen, the load falls, arid at a certain moment, failure occurs. Thus, for ductile metals, the ultimate strength characterises :, the resistance of the metal to considerable plastic deformation. In a tension test, in addition to strength indices’ such as 9,, 9,, 6,, and o,, ductility indices are also determined. They indicate the ability of the metal to deform in static tension tests without failure. * The reasons for the occurrenée of a yield step are not suifiently clear as yet. _. It 1s most frequently explained by the segregation on the grain boundaries of briftle components that impede grain deformation.™ The Mechanical Properties of Metals Ductility indices comprise: Retative elongation 8, which is the ratio of the extension in length of the specimen after fracture to its initial gauge length, expressed in per cent: 6 =a 100 per cent in which J, is the gauge length of the specimen after fracture. Relative reduction of area t which is the ratio of the reduction of cross-sectional area in the fractured specimen to the initial cross- sectional area, expressed in per cent: y= In determining a,, 0,, and @,, the corresponding loads P,, P,, and P, are divided by the initial cross-sectional area f of the specimen. The tension test diagrams shown in Fig. 51, a and & do not take into consideration the considerable reduction of area of ductile metals at stresses above the yield point and particularly upon necking down of the specimen. A more accurate relationship between specimen deformation and actual stresses is obtained from a true stress-strain diagram. This diagram is plotted in the coordinates true stresses (S)-true elongation (:). The true elongation e is the length increment of the specimen at each uistant referred to the actual length at that instant, instead of to the initial length: e=in( ! )=mn be Te Ie in which: » is the reduction of area of the “neck”, ? is the initial cross-sectional area, in sq mm, f, is the cross-sectional area in the process of deformation, in sq mm. A true stress-strain diagram is shown in Fig. 51, c. On this diagram the yield strength is denoted as S, and the iailure stress as Sy. The values of o, and S, are practically the same for all brittle metals which fracture without considerable plastic deformation. S, is always larger than , for ductile metals. It is evident from Fig. 51, ¢ that the true stress-strain diagram has two straight line sections corresponding to the ranges of elastic and plastic behaviour. Since the plastic deformation is much more important than the elastic deformation, the range of elastic behaviour is frequently considered as coinciding with the Y-axis; such a diagram is drawn as in Fig. 52. One of the most important values that characterise the properties of a metal in the range of plastic behaviour is the modulus of strain hardening D. This modulus is equal to the tangent of the angle that the straight line, in the plastic range of a true stress-strain curve, makes with the X-axis (Fig. 52). It indicates the capability of the metal to be strain hardened (hardened by plastic deformation). TheTension Tests 5 steeper the straight line in the plastic range of the curve, the higher the modulus of strain hardening, i.e., the higher the stress must be to obtain the required amount of plastic deformation. The larger D is, the higher the cold hardenability of the metal and the more it is subject to cold hardening in the process of plastic deformation. Strain § Fig. 52. True stress-strain dia- Fig. 53. Diagram showing the relation gram of the brittle and ductile state of solids to the temperature and rate of deformation (after A. F. Joffe) The ductility of a material depends upon the main values character- ising its mechanical properties: —6, eats Therefore, the larger a, and D are, the less the ductility will be. An increase in S, increases the ductility as well. The amount of work to carry a material to failure, or the toughness (see impact tests, Sec. 5-4), is equal to the area under a complete stress-strain curve. It may be expressed as a function of S,, 6, and D: sk—o} 2D a,= Hence, comparatively small variations in the values of S, and o,, which are in the second power in the equation, may substantially affect the toughness though they do not have a noticeable effect on the ductility. Mechanical properties depend on the test conditions. It has been established that lowering the temperature or increasing the rate of deformation considerably increases the resistance to piastic deformation (G,, 0,, as well as D) (Fig. 53), though they do not change% The Mechanical Properties of Metals the cohesive strength (see Sec. 3-1). Thus, at low temperatures (or high rates of deformation), metals and alloys, which are ductile at normal room temperatures (or low rates of deformation), may fail with a brittle fracture because the cohesivé strength S; is reached at stresses below the yield point. The higher the cohesive strength S,, the lower the temperature at which a metal or alloy will pass over from a ductile to a brittle state, i.e., the less it is susceptible to brittle racture. In tension tests, normal tension stresses are the more important since they are twice as large as the tangential stresses. Therefore in testing high-strength alloys having low ductility, compression and torsion tests are being more frequently employed. Here, the effect of normal stresses is considerably less than that of tangential stresses. Tangential stresses — Hi a Wormat tension stresses Fig. 54. Mechanically-stressed state diagram Fig. 55. Stress distribution in a showing the effect of the stressed state on the tension test of a notched f m mechanical properties: - men! pee 4, 2—penetration, scratching, 3—torsion, 4—tension, ‘O,—exial stress, O.—radial stress, and §-—tension of a notched specimen, /—solt metals, G,~-transverse stress, directed along Ogee leas oie. Beieadaeay TOMER le egg oF The diagram of the mechanically stressed state (Fig. 54) shows the walitative effect of the method of loading on the type of fracture Q paration or shear) obtained. tial Troax->Smar: i.e., if the tangential stresses are large In comparison to the normal stresses, the stressed state will be of the “mild” type and the metal will fracture by Shear which will be preceded by considerable plastic deformation. If, on the other hand, S...>>Tmax» then the metal will be in the “severe” stressed state and will iail by separation (brittle fracture) without noticeable plastic deformation. Notches, i.e., sharp changes in cross-section, have a large effect on the mechanical properties of metals. Notching in a specimen isHardness Tests 7 ilar, in effect, to actual machine components of intricate shape, h through holes, thread, etc., or to internal defects in a metal in form of various discontinuities (nonmetallic inclusions, graphite ‘east iron, cracks, etc.). As is evident from Fig. 55, a notch will Fause nonuniform distribution of stresses and thus lead to stress con- gntration. Nonuniform distribution of stresses, though they often jrease the resistance to plastic deformation (g,, o,), will always Bntribute to lowering the ductility of a metal (6, »). The effect. of B notch must be regarded among the factors that contribute to the tran- tion of a metal to the brittle state. A notch reduces the ultimate gth of -high-strength and brittle materials. The sharper the otch, the more its effect will be on the mechanical properties of a petal. Notch sensitivity, which is defined as the change in properties | a notched specimen, is one of the most important characteristics materials and is detérmined by special tests.* 8-3. Hardness Tests Hardness is the resistance of a metal to the penetration of another ‘der body which does not receive a permanent set. Hardness tests consist in measuring the resistance to plastic de- ation of layers of metal near the surface of the specimen. In the oeess of hardness determination when the metal is indented by special tip (steel ball or diamond cone), the tip first overcomes resistance of the metal to elastic deformation and then, a small jount of plastic deformation. Upon deeper indentation of the tip, overcomes large plastic deformation. Therefore, hardness is not an dependent characteristic of the mechanical properties. It determines the same properties as in other testing methods, for example, tension fests, but under different loading conditions. . This fact enables a celation to be established between the hardness jad ultimate tensile strength of ductile metals. For metals of. low uctility, in which the tensile strength is characterised by the cohesive strength and not by the resistance to considerable plastic deforma- ion, the relation between hardness and ultimate tensile strength is ot sufficiently reliable. : : It is necessary to distinguish; 1) macrohardness, which is the hard- ess of a material determined b: tesistance to plastic deformation na volume so large that the differences in actual hardness of its com- Eponent microvolumes are of no influence, and 2 microhardness, FWhich is the hardness of materials in microscopical ly small volumes ness of the separate structural components). For more detail see N. A. Shaposhnikov, Mechanical Testing of Medols, Mast- Moscow, 1964.78 The Mechanical Properties of Metals Hardness measurements for determining the properties of a machine component have found extensive application in the quality control of metals and metal products in all branches of industry due to the rapidity and simplicity of the tests and their nondestructive character. The most widely employed hardness testing methods are: Bail-indentation tests (Brinel! principle). This method consists in pressing a hardened steel ball under a constant load P into a specially prepared flat surface on the test specimen (Fig. 56). After re- l, a 2 y a — o ad a@ Fig. 86. Brinell hardness test: a—indentation having encircling ridge, 6—indentation having encircling depress ion moving the load, an indentation remains on the surface of the test specimen. If the area of the spherical surface in the indentation is denoted as F sq mm, the Brinell number Ban will be Bhna= £ kg/mm? Thus the Brinell number Bhn is the ratio of the load, applied to the ball during the test, to the area of the indentation produced by pressing the ball into the test specimen. If the indented area F is expressed-y the ball diameter D and the diameter of the indentation d, then the Brinell hardness number may be determined from the formula 2P Bha= =D D-VB= a kg/mm* The ball diameter and applied load are constant and are selected to suit the composition of the metal, its hardness, and the thickness of the test specimen (Table 2). The diameter of the indentation is measured with a special magnifying glass containing a scale graduated in tenths of a millimetre.Hardness Tests 79 Table 2 Brinelt Test Conditions (according to GOST 10241-40) Thickness of test speciinen, mm Ratio of load Materiai P to bi ame Steel, cast) up to 450) ows er 6 10 | 3,000 iron from 6 to 3} P=30D* 5 730 | 10 to 30° less than 3 2.5 187.5 Copper and} 31.8 to over 6 10 1,000 its alloys, 130 from 6 to 3) P= 10D 5 250 a magnesium less than 3 2.5 62.5 alloys, etc. Aluminium, | 8 to 35 over 6 10 250 babbitts from 6 to 3} P=25D* | 6 62.5 Cy less than 3 25 15.6 * For hardnesses op to Bha 140, the time of toad application is 30 sec; for harder materials — 10 see. In practice, the Brinell numbers corresponding to a given observed diameter of indentation d are taken from tables to simplify calculations. The following relationship exists between the Brinell hardness number and the ultimate strength of metals which form a neck in tension tests: o,=k Bhn in which k=0.36 for rolled steel and &=0.3 to 0.4 for steel castings. it is not advisable to apply Brinell tests to materials having a hardness which exceeds Bhn 450 since the ball may be easily deformed and ihis will introduce errors into the test cesults. The hardness of the ball should be at least 1.7 times higher than the test specimen to prevent permanent set in the ball. Determining hardness by the depth: of indentation of a diamond cone or small steel ball (Rockwell principle). This test, like the Brinell test, is based on the indentation a hard tip or indenter under a static load into the test specimen. In the Rockwell test the hardness is determined not by the diameter but by the depth of the indentation or impression. The indenter or “penetrator” in a Rockwell test may be either a conical-shaped diamond (called a “brale") with a 120° apex angle or a hardened steel ball 1.5875 mm ('/1,") in diameter. The brale is used for testing materials with a high hardness and the steel ball for softer materials. The brale and the ball are indented by two consecutive loads: the preliminary load P. (equal to 10 kg) and the additional load P which equals 90 kg for the ball (Scale 8) and 140 kg for80. The Mechanical Properties of Metals the brale (Scale C).* A brale with an applied load of 50 kg (Scale A) is employed for very hard materials and thin test specimens, The Rockwell superficial-hardness tester is applied for testing thin layers near the surface or thin specimens (nitrided steel, safety-razor blades, etc.). In this machine, the preliminary load is 3 kg and the total loads may be 15, 30, or 45 After foading, the additional load is removed. The Rockwell hardness number is the difference in depths of the indentations made by applying the total and preliminary loads, measured after removing ‘the additional load (Fig. 57). Fig. 57, Hardness determination with a diamond cone (braic) in a Rockwell test: By pen a ante pe i id. $-3—penetrs aff i a te Ps The Rockwell hardness number is read directly on the dial of the instrument having a scale graduated in hardness units which are determined by the formula Rt in which k Is e cmmstant equal to 0.2 for the brale and 0.26 for the A, Is the depth of penetration, in mm, of the brale (or ball) after removing the additional load (Fig. 57); A is the depth of penetration, in am, of the brale (or ball) under the preliminary load P. (Fig. 57); ¢é is the value of the scale divisions (0.002 mm). Rockwell U tests are are widely applied in industry due to the rapidity and i icity with may be fers formed, high accurac achi due to the small size of impressions produ The Rockwell hardness number may be converted into the Brineli "Determine Rardness by indent Determining by indentation of a diamond pyramid (Vickers principle **). ‘This testing method is used for determining hardness of * The total load fs the sum of the preliminary and additional toads, °° According to GOST 2090-45,Harden Tests ai Shusided et Specnens hee sen ie nerd , wi » etc. a ESS. test consists in forcing a tare based dfamond Yyrauid (with an iy 'angle of 136° between opposite faces) inte the groun: or even polished ‘ suriace to be tested. The hardness number is defined as the ratio of the F’ load in kilograms to the surface area of the indentation in square mil- ;! limetres. The latter is calculated by measuring the length of the dia- i onals of the indentations. The hardness number is determined from aP sing Pp VHN =—y— = 1.85445 in-which P'is the applied Joad (6, 10, 20, 90, 50, 100, or 120 kg); : a is the angle between opposite faces of the pyramid (@= == 136}; @ is the arithmetical mean, in mm, of the two diagonals . Measured after removing the load. ‘The load is selected in accordance with the size and hardness of the i specimen. The thinner the specimen, the less load required. On the her hand, the higher the load, the more accurate the results will be. , As a rule, the Vickers hardness number is determined from special tables in accordance with the measured value of d (length of the diagonal). Vickers and Brinell hardness numbers are expressed in the same : units and coincide for hardnesses up to about 400. At higher hardnes- , $08, the Vickers number is larger the Brinell number. > Microhardness. At the present time microhardness tests are widely ; used to determine the hardness of exceedingly thin layers, very small specimens, and even separate structural components of alloys, This method is based on combining a hardness tester, using a diamond : ‘potamid indenter at low loads, with a metallurgical microscope. Loads oS | to 200 g are applied. The microhardness number is determined from the formula . Pp H = 1.85445 in which P is the load, ing @ is the length of the diagonal, in microns. Specimens for microhardness tests must be prepared in the same manner as microsections. {t is advisable to use electropolishing as this removes the thin layer of cold-hardened metal, that is inevitable | in grinding the surface and which may affect the results of the test. The results obtained in measuring the microhardness of the. various. structural components in a nitrided layer are shown in Fig. 58 (see also page 265). . ;62 The Mechanical Properties of Metals Fig. 58. Determining microhardness of various phases in a nitrided layer, X 400 5-4. Impact Tests Static tension tests of unnotched specimens do not always reveal the susceptibility of a metal to brittie fracture. This important factor is determined in impact bending (or simply impact) tests. _Of all types of impact tests, the notched-bar tests are most extensively used (Fig. 59, a). Tests are conducted on a pendulum-type impact testing machine (Fig. 59, 6). The specimen is placed on its supports or anvils so that the blow of the striker is opposite the notch. The pendulum of weight Gis then raised to the height 4,, from where it is released to fracture the specimen and rise again to the height Ay. The energy required to fracture the specimen will equal A=G(h,—h,) kg-m.impact Tests 3 The heights of the pendulum before and after the blow may be expressed is the angles through which it is raised. In this case, the energy will be A=GiI (cos a,—cosa,) kg-m. The initial angle through which the pendulum is raised is a constant value and the upward swing after fracturing the specimen is 6 Me ek Fig. 59. A notched-bar test: a-notched-bar test specimen, b—pendulum-type Impact testing machine measured on a special circular scale on the machine. When the energy A, required to rupture the specimen, is known, the impa:i strength may be determined: Oy = + kg-m/em? in which f is the cross-sectional area of the specimen, in sq cm, at the fracture. Impact strength is a complex characteristic which takes into account both toughness and strength of a material. The main purpose of notched-bar tests is to study the simultaneous effect of stress concentration and high-velocity foad application. Impact tests are of the severest type and facilitate brittle fracture. Impact strength values cannot as yet be used for design caiculationsLy The Mechanical Properties of Metals but these tests are, as a rule, provided for in the specifications of carbon and alloy steels. ‘ in impact tests, the fracture may be either brittle or ductile. Brittle fracture is performed by ration and is not accompanied by noticeable plastic deformation. The fracture, in this case, will have a bright ‘anular or crystalline appearance (Fig. 60, 4). In ductile fracture, lhe surface will have a duil-grey fibrous appearance (Fig. 60, 6) and Fig. 60. Fracture in Impact test of steel, X 2: @—brittle fracture, b—ductile fracture rupture will be preceded by considerable plastic deformation. Fre- quently a compound fracture, with both crystalline and fibrous areas, Is Ol i. The type of fracture depends to a great extent on the test conditions. In accordance with the temperature, state of stress, and velocity of load application, many metals and alloys, as mentioned earlier (see page 76), may be transformed from the ductile to the brittle state. Quality tests in industrial plants are usually conducted at normal room temperature, under the assumption that the test conditions are not less severe than the operating conditions. It must be noted, however, that many metals, including iron and steel and also zinc, are cold-bsjttle or cold short. The tendency of a metal for transition from ductile to brittle behaviour upon tempe- Frature changes is determined by. the “serial” testing method in which impact tests are conducted on a series of specimens at several temperatures, each lower than the preceding one. Curves indicating the variation of impact strength with the test temperature are shown in Fig. 61. The temperature (/,,) at which the transition from ductile to brittle fracture takes place in a metal is called the cold-shortness threshold. Most often, a gradual transition from ductile to brittle behaviour is observed on the impact strength vs temperature curve (Fig. 61, 6), i.e., a critical temperature range ofimpact Tests 85 », brittleness is noted. In this case, N. N. Davidenkov s1 is that "a temperature corresponding to 0.4 a,,,,,(Fig. 61, 5) is accepted as the . transition temperature (f,,) of cold-shortness threshold. The higher the transition temperature, the more susceptible a metal is to cold ; shortness. Impact strength depends on the structure of a metal. Coarse grain } structure and precipitation of brittle layers at the grain boundaries do wot appreciably alter the mechanical properties in static tension but a ™ - | Impact strength, kg-m/en* Impact strength, hy-i/ car Test temperature, "C « é Fig. 61. Critical i of brittleness for steel, revealed by impact tests * Meat tempera! “fatter NA. ‘Shaposhnikev): by im {| @—sharp transition, b—gradual transition, tggtritical transition temperature (cold-sbortnets ’ threshold) ;. they substantially reduce the impact strength and raise the cold-shortness threshold. The shape of the notch and the size of the specimen greatly affect the impact strength. Therefore, to obtain comparable results, tests ; must always be conducted under the same conditions and on the same q of specimen. In conclusion, it may be noted that impact tests are most frequently t conducted for cold-short metals and alloys, steel, in particular, in ; which the relationship between the cohesive and shear strength varies ; Sharply upon changes in structure and test conditions. Impact tests on many steels enable the cohesive strength to be determined indirectly when this is impossible for static loading at temperatures. Impact tests are of little value for most wrought nonferrous metals (aluminium, coppers and many of their alloys) since there is no tran- - Sition to brittle behaviour at any temperature. In some cases, due to their high ductility, no fracture whatsoever is observed in ii of this group of metals subject to impact notched-bar tests. There is also no necessity of conducting impact tests for many cast alloys (cast © iton, cast aluminium and magnesium alloys) which show a brittle fracture in ordinary static tension tests.86 The Mechanical Properties of Metals 5-5. Fatigue Tests Most machine components and structural assemblages are subject to a variation, in both value and sign, in the applied loads (Fig. 62). Fig. 62. Examples of cyclic loading It has been established that metals subject to reversing stresses (com: »*ession (—a) and tension (+0)], repeated at definite intervals of time (Fig. 62), will fail at stresses which are not only lower than theultimate strength o,, but even lower than the yield point o,. Failure of the specimen will also occur after repeated variations only in the value of the stress even if its direction does not change. The failure of a metal under repeated pulsating or Teversing stresses is called fatigue. A typical fracture, containing two distinct zones (Fig. 63), is obtained as a result of fatigue failure. The first zone, called the fatigue zone, has a smooth surface and is gradually formed. Fig. 63. A fatigue failure First, a microcrack is formed in the weakest region of the cross- section. This gradually develops into a macroctack, After the crack has propaga to occupy a considerable of the cross-section, brittle fracture occurs in the remaining part. The fracture in this second zone is granular of crystalline in appearance as is typical for brittle fracture.Fai Tests a7 The exact nature of fatigue phenomena has not as yet been sufficiently explained but, probably, the chief reason lies in the nonuniform distribution of elastic and plastic deformation in micro- (or macro-) volumes of the metal. In selecting a metal for manufacturing a machine component it is necessary to take into consideration its resistance {> fatigue failure, ie its ability to withstand a large number of repeated varying siresses. The resistance of a metal to fatigue failure is characterised by its fatigue limit (endurance limit). The fatigue or endurance limit ¢, is defined as the maximum stress which a specimen can endure withou' failure when this stress is repeated for a specified number of cycles. This number of cycles is called the basis of the test. In accordance with GOST 2860-45, the fatigue or endurance limit is determined on the basis of 5 million cycles for steel specimens and on a basis of 20 million cycles for light cast alloys. Many experiments have proved that if a metal withstands the above-indicated number of cycles without failure, it can withstand this same stress for a considerably larger number of cycles. The fatigue limit is usually determined { by subjecting a rotating specimen to com- Vv P letely reversed flexural loads which develop reversing stresses (tension-compression) in a symmetrical cycle (called an alternating cycle). A schematic diagram of a fatigue test, with fiexural cantilever loading of a rotating specimen, is shown in Fig. 64. At least six specimens must be tested to determine the fatigue 1i- mit. The first specimen is tested at astress (01 kg/mm’) equal to0.60, for steel and the number of cycles N, at which failure occurs, is determined. The stresses o,, 0,, etc., for the second and subsequent specimens are increased or reduced by 2 or 4\kg/mm?, depending on the number of cycles which caused the failure of the first specimen. The results of the tests are plotted on a diagram having the coordinates stress (o kg/mm) vs number of cycles (4) in either rectilinear or logarithmic scales. Such a curve is shown in Fig. 65. The horizontal section is a straight line and is the maximum stress at which failure will not take place after an infinite number of loading cycles. This value is the fatigue limit, which depends to a large extent on the size of the specimen, stress concentrators (fillets, grooves, etc.), surface finish, corrosion, and upon other factors. An increase in the size of the specimen will reduce the fatigue limit. The presence of stress concentrators, such as notches (sharp changes in cross-section, grooves, etc.), on the specimen will sharply reduce the fatigue limit. The finer the surface finish of the specimen (or machine component), Fig. 64. Principle of a fatigue test8 The Mechanical Metals the higher the fatigue limit will be. Corrosion substantially reduces the fatigue limit while surface hardening of machine component (carburising, nitriding, case hardening, etc.) raises the fatigue limit and considerably reduces the harmfu " effect of stress concentrators (see page 76). 1, 5 Fig: 6. Saeed Ui eermene, between the ile m8 pea a imetal i a (large Ua, J—crankshalt “ee com tenon ami Pe aa It must be noted, in conclusion, that fatigue or endurance tests do not provide sufficient information concerning the strength of machine com ts since they do not fully reproduce the forces and treatment that the metal is subject to in manufacturing the “in pole wel bsequent ation in a machine, is principle is well illustrated in Fig. 66. Therefore, only tests on the components themselves may properly determinefatigue limits. In the last years, fatigue tests are conducted on the ine com its or on specimens of a shape and size near to these ents, Tinese are called fullsize fatigue tests. 5-6. Tests at Elevated Temperatures ¢ Many com its of modern machinery operate under high-tempe- ure conditions. An increase in temperature has a large effect on }] mechanical properties. Research has shown that elevated tempera- es reduce the modulus of elasticity (by decreasing interatomic bind- 2), yield point, ultimate si th, and special the modulus of istrain Hardening in the process of plastic d ition (see page 74). [; If a metal is heated above the recrystallisation temperature, the yield point and ultimate strength may be reduced several times Fwhile the modulus of strain hardening may be reduced tens and hund- Ereds of times. . At elevated temperatures, the mechanical properties of metals are result of the simultaneous processes of strain hardening, due to plastic jeformation, and the softening effect of recovery and recry- ‘atallisation. _ Continuous plastic deformation takes place in metals and alloys, constant loads at elevated temperatures, and it may lead to The slow continuous plastic deformation of metals under a constant doad (especially at high temperatures) is called creep. » Special tests are conducted to determine the strength of metals Pat tievated temperatures. The following chief types of high temperature tests are most frequently conducted: E'- Short-lime tension tests are similar to tension tests at room temperatures except that the specimen is placed in a furnace. The aim of these tests is to determine the strength and ductility at a specified p temperature. Creep tests. The purpose of these tests is to determine the creep limit, i.e., the maximum stress that may be applied for a long period t of time at a given temperature if the rate of creep (over this given period of time) is not to exceed .a specified value. . : A creep testing installation is illustrated in Fig. 67. The specimen is placed in an electric furnace where it is heated to the given tempe- { rature and is constantly subject to a load applied by a lever and weights. The strain in the specimen is measured by an optical exten- someter with an accuracy of 0.001 mm. Four or five specimens are tested at each temperature under different loads and elongation vs + time curves (creep curves) are plotted for each specimen. ; The section Oa on creep curves (Fig. 68) corresponds to elastic "and plastic deformation due to the momentary application of the90 The Mechanical Properties of Metals load. In the following section ab, the metal is elongated at a nonuniform (decreasing) rate and irr section dc the rate of creep is uniform. The creep limit may be characterised, for example, by the stress which produces a total elongation of 1 cent in 300 hrs. On the creep curves, this will correspond to the stress 0. If the specifications GEES GE Fig. 67. Install ition for creep Fig. 68. Creep curves ing: f—spectmen, 2— S—tumsce, require that the creep limit be determi- ie, tance Smiteees: ned for an elongation not exceeding 1 for strain “meaturement er cent in 100 hrs, the stress value will e 0, (Fig. 68). , Only the creep rate for a steady process is taken into consideration for machine components that are designed for continuous operation at elevated temperatures. Limiting conditiorts are specified for such com> onents. In practice, the stresses are most frequently determined lor @ total elongation of 1 per cent in 1,000, 10,000, and 100,000 hrs. ese correspond to average creep rates of 10~*, 10-‘, and 10~° per cent per hour, respectively. Creep-rupture tests are creep tests which are carried on to the failure of the specimen. In these tests, the time to fracture the specimen, at a Given temperature and stress, is measured. Often permanent elongation and reduction of cross-sectional area are also measured. 5-7. Fabrication Tests Simple tests are often used to check the suitability of metals for particular fabrication techniques. Their purpose is to determine whether a metal can withstand the deformation it will be subject to during operation or during cold or hot working procedures employed in manulacturing.Fabrication Tests ot The results of fabrication tests are usually evaluated by the appea- tance of the metal after the test; the absence of tears, cracks, lamination, or fracture shows that the metal has passed the test. Cupping tests are used to check the suitability of thin sheet metal for cold Pressing. drawing, and similar forming operations requiring plastic deformation. Bending tests (on sheets, bars, various sections, etc.) also serve to determine the capability of a metal for piastic deformation and to reveal flaws. These tests are classified as bending through a definite angle, bending about a mandrel until both ends of the specimen are parallel to each other, and folding. The method required may be indicated in the specifications. Special machines, presses or simply vises with rounded jaws are used for bending tests. Bending-and-unbending tests are conducted to ascertain the capability of a metal (wire, strip, or sheet material) for withstandiic considerable plastic deformation. Tests are conducted in the cod ate only. Wire twisting tests are also of the cold type. The specimen is considered to have passed the test if, after twisting (or untwisting), no lamination, cracks, or other defects are revealed. Fabrication ‘tests on pipe include: 1) expanding tests, 2) reducing tests, 3) flanging tests, and 4) flattening tests.Chapier 6 BINARY ALLOYS. CONSTITUTION AND EQUILIBRIUM DIAGRAMS . 1. General Principles so Transtormations in Most combinations of metals have the property of unlimited mutual solubility in the liquid state. Certain metals, for example, lead and iron, are almost completely insoluble in each other and separate according to their specific gravity in the liquid state. Complete insolubility is rarely encountered. Limit- ed solubility in the liquid state is much more Frequent. In such a case, a homogeneous liquid solution will be obtained if the amount of metal B added to metal A does not exceed its maximum solubility in metal A at the given temperature. Ef, on the other hand, the amount of added metal B exceeds its maximum solubility in A, the liquid will separate into two layers. These layers will comprise saturated solutions of B in A and of A in B. With few exceptions (Fe-Cu, Ni-Ag, Cu-Cr, and others), limited solubitity in the liquid state is characteristic of metals whose melting points and atomic volumes sharply differ (Cu-Pb, Zn-Pb, Cr-Sn, etc.). . When a metal is alloyed with another metal or with a nonmetal, chemical interaction occurs between the atoms of the elements forming the solution. The valence electrons of the alloying elements, weakly bound to their nuclei, do not belong to the separate atoms but are shared by the whole aggregate, as in the solid phase, comprising a single system for the whole liquid solution. The distribution in a liquid solution of atoms (ions) of elements constituting an alloy depends: a the relation of the interacting forces between like and unlike atoms. : The components of an alloy may either be dissolved one in the other in the solid state to form solid solutions or chemical compounds of, after solidifying, they may produce mechanical mixtures of the two components A and B, solid solutions, etc. In the formation of a Solid solution, atoms of the dissolved component (the solute) will substitute for a part of the atoms in the lattice of the solvent or they will be accommodated in the interstices (see Sec. 6-4). In the majority of cases, the com its of an alloy have limited mutual solubility in the solid state. y metals, however, are mutually soluble to an unlimited extent.General Principles of Phase Transformations in Alloys % '. _ If components are present in an amount exceeding the solubility limit, the surphs metal forms an independent phase comprising either a saturated solid solution or a chemical compound or, finally, " orystals of one of the components. A chemical compound differs ~ from a solid solution in that it often has its own crystal lattice and in many cases a definite quantitative relation of the components is required for its formation. : The transition of an alloy from the liquid to the solid state is asso- }, . clated with the conversion of the system to a state of less free energy. '. As in the solidification of pure metals, this transition will be initiated only if the alloy is supercooled to some extent and will then proceed }.. by nucleation and subsequent growth of the nuclei.as dendrites or as holohedral crystals. In actual conditions, the crystals, produced in m. ‘the solidification of metal alloys, are usually dendritic. When alloys solidify, the newly-formed i ases differ substantially in composition from the initial liquid solution. Therefore, to form a iy stable nucleus or embryo, it will not be sufficient to have fluctuations 1 of energy as in a single-component system; there must also be fluctua- fons concentration which correspond to zones not less than critical Hin size. . ; i. Concentration fluctuations are chance, tentative deviations of the - chemical composition from its average analysis and occur in separate } small volumes of the liquid (or solid) solution. The reason for the wccurrence of fluctuations of concentration lies in the diffusion of F. gions in a liquid (or solid) solution due to their thermal vibrations. Nuclei of the new phase may form only in those microvolumes of the liquid phase where, due to fluctuations, the concentration conforms to i the composition of the solidifying phase. H then, such fluctuations in Lin and concentration correspond to microvolumes of a size not less than the critical value, a stable embryo will be formed, one that is ‘apable of subsequent growth. ie rate of crystal growth in solutions is considerably less than in B crystals. The reason is that crystal growth (in a solution) is not B -@hly associated with the formation of two-dimensional nuclei on the E -etystal faces, as in pure metals, but also with atomic diffusion of the ;. mponents in the liquid solution. The more the difference in concent- tation between the liquid solution and the crystals into which it ;. #olidifies, the slower the rate of crystal growth will be. After an alloy has solidified, structural changes may occur that fafe caused by polymorphic (allotropic) transformations and by the decomposition of the solid solution due to changes in mutual solu- . bility of the components in the solid state. - ‘. ' Transformations in the solid state are of a crystallisation type only f and proceed by nucleation of the new phases and subsequent \. ; A distinguishing feature of solid state transformations is that F the principle of structural and size conformity is usually observedve Binary Alloys. Constitution and Equilibrium Diagrams in nucleation of the new phase. This means that the nucleus of the new phase is oriented so that it joins the initial phase on definite crystallographic planes which more closely match each other as to atomic arrangement and interatomic distances. As long as the lattices on the boundary of the old and new phases match along definite crystallographic planes (Fig. 69) or, as it is Ola nase Fig. 69. Schematic diagrams 0’ ihe relationship between the fattices of the old and new’ phases: a (ranaitton of one lattice into the of! crystal grow! SSeereat Nate. Bi Tuck “of Continultys ta well Bo dhscraet “atthe” tnkerfoee, tops coherent growth (afier G.V. Kurdyumav) called, the lattices are coherent, the growth of the new phase proceeds rapidly since the atoms pass orderly over small distances. The nucleation of the new phase, however, will create elastic energy due, for example, to the difference in specific volume of the initial and new phases. At a certain moment, the magnitude of the elastic energy may exceed the elastic limit of the medium. This will result in shear deformation and loss of coherency. Further coherent growth will be impossible. It must be rioted that coherent growth may also be violated by the thermal vibrations of the atoms As soon as the lattices no longer match on the boundary between the phases, further growth of the nucleus proceeds at a much slower rate since the atoms pass over from one phase to the other in a disorderly manner. At low degrees of supercooling, coherent growth rapidly ceases and the transformation will have a disorderly character. Upon high degrees of supercooling, due to rapid cooling, in many cases the new phase will be subject to coherent growth. Violation of coherency will stop the growth of the new phase since the diffusion processes, required for the passage of atoms from one phase to the other, become impossible at low temperatures. Nuclei of the new phase are easily formed on grain boundaries and around various inclusions. This is explained by the fact that grain boundaries, as well as the surface of inclusions, have a layer of atoms with increased free energy where the new phase may appear.The Phase Rule and Equilibrium Diagrams 95 Therefore, the smaller the grains, the more rapidly transformations will proceed. The formation and growth of nuclei of a new phase in the solid state are hampered by the difficulty in obtaining the required fluctuations of concentration. This difficulty, in turn, is due to the low rate of diffusion and the influence of elastic energy. This explains why phase transformations proceed at a high rate only upon considerable supercooling of the initial phase. 6-2. The Phase Rule and Equilibrium Diagrams The solidification of metal alloys is clearly demonstrated by means of equilibrium diagrams which are convenient graphic representations of changes in state due to variations in temperature and concentration. Equilibrium diagrams, atso called constitutional and phase diagrams, enable the phase content of the alloy to be determined at any temperature and composition. They enable the phase transformations to be followed in heating or cooling the alloy under equilibrium conditions, i.e., when all processes in the given system are reversible. This means that changes occurring in a system as a result of processes proceedii in one direction are fully compensated by changes due to the reversa of the process in the system. A system is here understood as the whole complex of phases of one or several components at different pressures and compositions. The components are the substances {chemical elements or compounds) whose presence is necessary and sufficient to make up a system. For example, a pure metal comprises a one-component system, an alloy of two metals is a two-component (binary) system, etc. A phase is a physically and chemically homogeneous portion of @ system, separated from the other portions by a surface, the interface. For instance, a homogeneous liquid solution is a singie-phase system; a mixture of crystals of two types, differing in composition and structure and separated by an interface, or the coexistence of the liquid alloy and its crystals comprise two-phase systems. All changes which take place in a system consisting of several phases, in accordance with external conditions (temperature and pressure), conform to the so-called phase rule. The phase rule estab- lishes the relationship between the number of degrees of freedom, the number of components, and the number of phases. It is expressed mathematically as follows *: F=Cta—P * For a thermodynamical derivation of the phase rute see: N. M. Wittorf, The Science of Alloys, Moscow, 1927; and A. A. Bochvar, Physical Mefatiurgy, Metallur- gizdat, Moscow, 1956.6 Binary Alloys. Constitution and Ei Di in which F is the number of degrees of freedom in the system (the number of variable factors); C is the number of components in the system; P is the number of phases in equilibrium; nis the number external factors (fot example, tem- . perature and pressure). The number of degrees of freedom is the quantity of independent external or internal variable factors (temperature, pressure, and concentration) which may be altered without causing the disappearance of a phase or the formation of a new phase in the system. In studying chemical equilibrium, temperature and pressure are regarded as external factors determining the state of the system. The effects of pressure may be neglected in appl the phase rule to metal systems leaving only one variable external factor—temperature. The equation will then be: F=ol+1+P All internal and external factors (concentration and temperature, respectively) have definite values in a system that is in equilibrium. ince the degrees of freedom cannot be less than zero C—P+1>0, then PeC+1, ie., the number of phases in a system cannot exceed the number of components plus one. Therefore, no more than three phases may be in equilibrium in a binary system, no more than four in a ternary system, etc. In cases when the maximum possible number of phases are in equilibrium, the number of degrees of freedom equals zero (F=0). This is called nonvariant equilibrium. As in nonvariant equilibrium may exist only under entirely definite conditions: at a fant temperature and at a definite composition of all phases involved. A pure metal at the solidification temperature, for example, is a gne-component system consisting of two phases of identical compo- ion F=i+l—2=0 This means that it is a nonvariant system. In this case the tem- ature cannot be selected or changed arbitrarily. There is only one lemperature at which the system is in equilibrium. This is the meltingThe Phase Rule and Equitibrium Di solidifying) point for the given metal. If the number of phases is than the maximum possible number by one, the number of degrees freedom will also increase by one (F=1). Such a system is said to be ovari . ‘An alloy of two metals, for example, is a two-component, two- system, in the general case, at the beginning of solidification. therefore, F=1 in this case. A system with F=2 is divariant, There- pre, the system may be in equilibrium at different temperatures and centrations. A number of examples of practical application of the phase rule studying the phase diagrams of two-component alloys (C==2) given oo vy For practical purposes it is essential to know the te state Band constitution of alloys depending on the temperature concent- ttation. If the state of an alloy of given concentration is known at fany temperature, it is ible to predict its properties, its capabilit for heat-treatment plastic deformation (forging, stamping, roli- king, etc.), its behaviour in casting, its suitability for various operating Econditions, etc. All transformations occurring in an alloy and depend- Hing on temperature and concentration can be Clearly represented by Fequilibrium diagrams, These diagrams are plotted wi concentra- Etion as the abscissa and the temperature as the ordinate. Thermal analysis is employed as the basis in plotting equilibrium ams and especially for determining freezing points. By this method, extensively developed by N. S. Kurnakov, cooling curves rare plotted first. Then the eritical points are determined by the abrupt Einflexion of the curves due to thermal effects in transformations. At the present time, all methods of physico-chemical analysis are Hemployed in studying solid-state transformations. They include: ‘micrography, X-ray, dilatometric and magnetic analyses. A complete equilibrium diagram may be drawn after experimentally determining ‘the critical points of a series of alloys. - - In principle, equilibrium diagrams represent quite definite types of equilibrium: complete or partial solubility, presence or absence of chemical compounds, ete., and may be drawn on the basis of theoretical considerations. In some cases, actual equilibrium diagrams of alloys almost completely conform to these theoretical types of diagrams. More fre: ly, q they are a combination of several types of diagram. 4 all ilibrium diagrams are valid only under the condition that -ocesses, which. may occur in the given system, are of the equi- OA oe ypes of two- t or binary equilibrium di j e mai es component or equilibrium diagrams } are considered | in the succeeding sections. , . : : 713786 Binary Alloys Constitution and Equilibrium Diagrams 6-3. Equilibrium Diagram of a Binary System in Which the ints Form a Mechanical Mixture of Crystais In the Solid State and Are Completely Mutually Soluble in. the Liquid State After solidification the alloy will consist, in this case, of a mechanical mixture of the two initial components. * Metals, which form alloys of this type, retain their crystal lattices. This has been proved by radiographs which reveal lines corresponding to both components. Doman 2 ow ow nao noe Ne So component a Fig. 70. Constructing an equilibrium diagram for a system of components which are completely mutually soluble in the liquid state and form a mechanical mixture upon freezing We shall consider equilibrium diagrams of the first type usi alloys composed of metals with the generalised symbols A and as an example. We shall first melt metal A and then cool it slowly, measuring the temperature after equal intervals of time. As is evident in Fig. 70, the temperature of the pure metal A falls gradually to ¢, at which metal A solidifies. A horizontal step appears on the curve because, according to the phase rule, the solidification of a pure metal proceeds at a constant temperature (F=1+-1—2=0). When the metal has solidified completely the uniform drop in temperature is resumed. . “ Strictly speaking, complete mutua! solubility of two metals never exists. Aconsideration of this diagram, however, is of methodological value as it facilitates the study of the subsequent types of equilibrium diagrams,Binary Systems in Which a Mechanical Mixture is Formed 9 If, now, an alloy of 80 per cent component A and 20 per cent coniponent B is taken instead of the pure metal, the temperature will fall uniformly to ¢,. First we note the inflexion in the cooling curve at ¢, (first critical point). After this point, the rate of cooling is markedly reduced due to the evolution of the latent heat of fusion resulting from the freezing out of a definite number of component A crystals. lidification proceeds at the new rate until temperature ¢, is reached. At temperature ¢,, the fall in. temperature ceases (the horizontal step on the curve). This corresponds to the completion of solidification of the alloy at a constant temperatur2 (second critical point). In cooling an alloy of 60 per cent A and M0 per cent B, it will remain liquid until it reaches temperature ¢,. At 4, there is a single long temperature stop associated with solidification of the whole alloy at constant temperature. in the solidification of alloys containing 20 per cent A and 80 per cent B, crystals of component B begin to freeze out of the liquid phase at temperature ¢’, The alloy completely solidifies at temperature ¢,. The cooling curves indicate that ail alloys of the A-B system solidify completely at the constant temperature ¢,. From this it can be assumed that the part of the liquid alloy which solidifies at the constant temperature f, is also of constant composition. This composition corresponds to the alloy consisting of 60 per cent A and 40 per cent B which completely solidifies at temperature ¢,. In all other compositions of the alloy, the metal that solidifies first is surplus in reference to the composition containing 40 per cent B. Solidification of the surplus metal proceeds until the concentration of the remaining liquid part of the alloy reaches a composition of 40 per cent & and 60 per cent A. In alloys of this composition, the components A and B solidify at the same time at the constant temperature ¢,. This occurs because, according to the phase rule, there is nonvariant equilibrium when three phases (one liquid and two solid phases, A and B) exist simultaneously in a two component system, i. e., F=-2+1—3=0. The surplus component (A or 8) freezes out of the liquid alloy in a temperature interval since the system has one degree of freedom: Fo2+1—2=1, Alloys in which the components solidify simultaneously at a constant temperature, the lowest for the given system, are cailed eutectic alloys. A mechanical mixture of two (or tore) hases, witich solidify simultaneously from the liquid alloy, is called a eutectic. * * From the Greek word eutekios which means: eashly melted. 7100 Binary Alloys, Constitution ond Equilibrium Diagrams If all the critical points that have been determined are plotted on a diagram with the compositions (concentrations) used as ab- scissas and the temperatures as ordinates (Fig. 70) and then if all these points of the same type (representing the same processes) are connected by smooth curves, the equilibrium diagram of the system of alloys A and B will be obtained. The alloys begin to solidify at temperatures which form the line ach (Fig. 71) called the liquidus. As indicated above, pure metal A Liquid Phase 8 crgstais ey} a Acrystais BS erystais sutectie (+8) eutectic (A+B) o 0 20 30 40 50 60 1 4 Yecomponent B . TL, Rquilibrlum diegram of a system of components A and B which ig 71 Betttaly soluble te the iqaid and incleble Ws doe'solid stats? ins to solidify out of the liquid alloy al the line eet B al Yhe line ob. q y atone ae and pure t The comple fe solidification of any alloy takes place at the constant emperature ich, on iagram, is represented by the horizontal line dee (Fig. 71). This line is calied the solidus and, inthe ‘given case, the fatectio rature line, A . point ¢. components, A and B, solldify simultaneously t Sows a scar phe dren, on le eu! ie alloy per cent B), as is clear in the diagram, solidifies at a constant temperature ¢, to produce: the eutectic which consists of crystals of the components A and B. When the eutectic temperature is reached, the liquid alloy is-saturated by both compo-Binary Systems in Which Mechanical Mixture is Formed 101 nents at the same time. Solidification begins only after a certain degree of supercooling is reached, when the solution becomes su- F. persaturated by both components and freezing of the eutectic becomes thermodynamically possible. Temperature, °C ! | Aer sélacs | eutedtte| (ang) a 0 20 30 0 50 ” Fig 72. Portion of the equilibrium diagram for the system of components A and B 4g (hypoeutectic alloys) . First, crystals of one component (for example, component A) appear FP and begin to grow. The liquid surrounding these crystals (A) becomes ficher in the second component (8) whose crystals begin to freeze out as a result. Next, the liquid adjacent to the B crystals becomes richer in metal A whose crystals again begin to freeze out. . Thus, by alternate supersaturation of the liquid by components & A and B, the eutectic embryo is formed. This nucleus continues to 7 since new atoms pass over from the liquid to the surface of crys- etals A and B. F’ ; This, a continuous diffusion process takes place in the liquid during _ eutectic solidification. In this process, atoms of component A conti- ~ _ tuously approach A crystals and atoms of component B approach B ks. The solid phase, first formed in the liquid, is called the primary Phase. Crystals of the primary phase may have various forms. They P May be dendritic, lamellar, or spheroidal (globular).loz Binary Alloys. Constitution and Equilibrium Diagrams The secondary phase fills the spaces between the crystals or between the branches of the dendrites of the primary phase. We may now consider the solidification in certain alloys of the system formed by components A and B (Figs. 72 and 73). An alloy containing 20 per cent 8, begins to solidify when it reaches temperature ¢, (with a certain degree of supercooling, of course) (Fig. 72) phere crystals of : compdnent 4 fin to se- ‘ate from the liquid al- or As the temperature falls, component A continues to solidify so that the composition of the re- alloy “thanges towards alloy Fy higher B content. weno Fig. 74. Equilibetum, diagram for Pb-Sb alloys The composition of the liquid phase that is in equilibrium with the crystals of component A may be determined for any temperature between the liquidus and solidus (for example, ¢, and 7,) by drawing lines through the temperatures parallel with the axis of concentrations (Fig. 72). The intersections of these lines with the liquidus {n and m) indicate the compositions of the liquid phase at the given peratures. It is evident from Fig. 72, that the composition of the liquid portion of the alloy varies along the liquidus, when theBinary Systems in Which a Mechanical Mixture is Formed 103 temperature is lowered, approaching the eutectic concentration of 40 per cent B (point c). At the temperature 4, the liquid portion of the alloy solidifies at constant temperature with simultaneous freezing of components A and B which form the eutectic mixture. After complete solidification, the alloy will contain crystals of component A, which initially separated in the temperature interval from f, to 4, and the eutectic, produced upon final solidification of the alloy at temperature ¢,. Alloys containing less than 40 per cent B are located to the left of point ¢ on the diagrams (Figs. 7] and 72)'and are called hypoeutectic alloys. After solidifying, all hypoeutectic alloys have a structure containing the eutectic and surplus component A (Figs. 71 and 72). The nearer an alloy is in composition to the eutectic alloy (40 per cent B), ihe more eutectic and less surplus crystals of component A it will contain. The solidification of an alloy containing 80 per cent B (Fig. 73) begins at temperature ¢, where surplus crystals of metal B start to precipitate from the liquid alloy. Lowering the temperature will change the composition of the remaining liquid portion of the alloy along the curve bc (liquidus) due to further separation of B crystals, At the temperature fy, for example, the composition is determined by point m and at temperature 1,, by point m. At temperature ¢,, the liquid alloy will reach the eutectic composition (40 per cent B+ 60 per cent A) and will solidify to form the eutectic mixture of A and B crystals. AI alloys located on the diagram to the right of point ¢ (containing more than 40 per cent B) are called Aypereutectic alloys. After solidifying, they will consist of crystals of the primary component B surrcunded by the eutectic. The nearer the concentration of the alloy approaches the eutectic composition, the more eutectic and less crystals a surplus component B will it contain. The equilibrium diagram enables the quantitative ratio of the phases to be determined at any point during solidification. The relative amounts of the liquid and solid phases present at temperature fs, for example, in a hypereutectic alloy containing 80 pet cent B (Fig. 73) are determined by the lever rule. This rule states that the ratio of the amount by weight of the liquid phase to that of the solid phase is inversely proportional to the corresponding sections of the straight line nt,k: Ni hase yk saiid phase” phase ~ ign Fig. 74 illustrates the actual equilibrium diagram for the system : Sb-Pb.* It closely resembles the typical diagram described above. * For the sake of simplicity, it has been assumed that Pb and Sb do not form solid solutions with each uther. Actually these metals exhibit slight mutua! solubility in the solid state.104 Binary Alloys. Constitution and Equitib rium Diagrams Fig. 75. Microstructure of Pb-Sb alloys, x 250: aaphypoeutectic alloy of 96 per cent Pb and 4 per tent Sb, lead erystals (darker) and eutectic (Pb Sb): '— hyporutectic alloy of 94 per cent FY and 6 per cent Sb; lead crystals darker) and eutectic; e—eutectic alloy, 67 per cent Pb and 13 per cent Sb, eutectic (darker areas are lead and ilghter areas are antimony); d—hypereulectic alloy of 70 per cent Pb and 30 per cent Sb, antimony crystals (lighter) surrounded by eutectic (Pb+Sb); ¢—hypereutectic alloy of 26 pet cent. Pb and 74 per cent Sb, antimony crystals and eutectic; {—hypereutectic allay of 95 per cent Sb and 5 per cent Pb, antimony crystals and eutecticBinary Systems in Which a Mechanical Mixture is Formed 105 Photomicrographs of Pb-Sb alloys, shown in Pig. 75, illustrate their structure. After solidification, these alloys form a simple mechanical mixture of the components. A Phenomenon known as gravity segregation is frequently observed in the solidification of alloys whose behaviour conforms to the first type of equilibrium diagram. This segregation (or liquation) is also evident in Pb-Sb alloys. The Pb or Sb crystals that precipitate in the process of solidification differ considerably in specific Fig, 76. Microstructure of the Pb-Sb alloy with 20 per cent Sb, X 250 (after Turkin and Rumyantsev): a—upper part of the ingot showing searegation of antimony crystals (lighter erystatsi; b--lower part of the ingot containing almost "pare eutectic (Pb+ Sb weight from the remaining liquid portion of the alloy and, consequently, they either rise to the surface (Sb) or sink to the bottom (Pb). As a result, the ingot will not be homogeneous in composition. The photomicrographs in Fig. 76 show the structure of microsections cut out of the upper and lower parts of an ingot (Fig. 76, a and 6). The alloy, in this case, contains 20 per cent Sb and 80 per cent Pb. These photomicrographs clearly show that the upper part of the ingot is rich in antimony due to segregation while the lower part is practically free of surplus antimony crystals and consists of the eutectic only. A number of measures are employed to prevent gravity segregation. They include: rapid cooling, stirring of the liquid alloy, and adding a third component which freezes first in the form of branching dendrites that prevent separation of the alloy into layers by purely mechanical means. The properties of alloys, consisting of a mixture of two phases, depend on the properties of the phases formed in the alloy. N.S. Kur- nakov (1860-1941) first showed the definite relationship between106 Binary Alloys. Constitution and Equilibrium Diagrams the composition, type of equilibrium diagram, and properties of an alloy. ha simple mechanical mixture is formed under conditions of low degrees of supercooling (Fig. 77), low temperatures, and different structural state of the alloys, certain of their properties will vary along a straight line. This means that they will be the average of the properties possessed by the initial components (or component phases). If the phases comprising the mixture form fine crystals, as in eutectic alloys, the strength and hardness are increased. In this case, the linear relationship between pro- erties and composition does not hold ttre, Oa of a mponen Mutually Soluble in Both the Liquid and Solid States General features of solid solutions. Most metals form homogeneous liquid solutions in the liquid state. In their transformation to a solid crystalline state, the homo- Fi geneit of many alloys, is retained. Conse . quently, their solubility is also retained. % component § * Said phases in which the ratio of the rie Tick and mbacitien components may vary without violating er ioys comprising Pimple the homogeneity are called solid solu- mechanical mixture of the fons. . ‘components In a solid solution, the different types of atoms of the components in the alloy form a common crystal lattice. The component, whose lattice is retained, is called the solvent. Solid solutions are classified as substitutional (Fig. 78, a) and interstitial (Fig. 78, 6). In a substitutional solid solution, a part of the atoms at the points of the solvent crystal lattice have been replaced by solute atoms (Fig. 78, a). - Since the sizes of the solvent and solute atoms differ, the crystal lattice is always distorted and the lattice constants are varied (Fig. 79) in a substitutional solid solution. As a rule, if the solute atom has a larger radius than that of the solvent atom, the lattice constant will be increased and vice versa. In interstitial solid solutions, the atoms of the solute metal occupy the vacant positions or interstices between the atoms of the sol- Temperature Properties (Bhi, 6,)Systems with Mutually Soluble Componenis 17 vent. Interstitial solid solutions may form only if the diameter of the solute atom is small and may be accommodated in the interstices of the solvent crystal lattice. Thus, solid solutions of this type are only found when elements with a small atom radius such as carbon, nitrogen, and hydrogen are dissolved in metal (for example, iron). An interstitial solid solution always increases the lattice constant of the solvent. a @ sotvent atom O Solute atom Fig. 78. Crystal lattice of a solid solution: G—aubstitutional, b—Interstittal Fig. 79. Distortion of the crystal lattice in substitutiona] solid solutions The metailic type of bond is found in ali metal-base solid solutions without regard for the type of bond peculiar to the solute in the solid state. Therefore, nonmetallic atoms (C, N, Si, ete.) do not change the metallic bond of metallic solid solutions. In 4 number of metal systems (for example Cu-Au, Fe-Al, Fe-Si, Ni-Mn, etc.) the atoms of both components occupy entirely definite positions in the solid solution lattice (Fig. 80) if the alloy has been slowly cooled or has been heated for a tong time at a definite tem-108 Binary Alloys. Constitution and Equilibrium Diagrams ature. To distinguish such solid solutions from those in which the atoms of the components assume statistically sandom repiace- ment positions (Fig. 78), the former are called ordered solid solutions. This new order of symmetry superimposed on the solvent lattice results in a structure called a “superlattice”. Ordered solid solutions usually occur when the ratios of the alloyed components in Aw Btomic per cent are equal to whole numbers: 1:1, 1:3, etc. Inthis case, the formula of a chemical compound, for example, Cu,Au (Fig. 80), Fe,Al, etc., may be assigned to the ordered solid solution. * The formation of ordered solid solutions is accompanied by sharp alterations in the properties of the alloy. Resistivity falls sharply, magnetic properties change, the alloy becomes Fig. 80. Unit cell of Very hard and brittle, etc. Ordered solid solu- an ordered solid solutions may be regarded as intermediate pha- tion (CuAu) ses between solid solutions and chemical compounds. Ordered arrangement of the atoms of both components in the lattice and a substantial change in properties are also characteristic of chenical compounds. Ordered solid solutions differ from chemicai compounds in that the solvent lattice is retained and in that a di- sordered solid solution is gradually formed upon heating without altering the type of lattice. A solid solution, observed under an optical microscope, can be seed fo consist of homogeneous crystalline grains (Fig. 36) as in pure metals. Substitutional solid solutions may be limited and unlimited or continuous, i. e., they may occur only in a limited region of alloy compositions or at any composition. a. Equitibrium Diagrams of Systems Whose Components Have Unlimited Solubility in Both the Liquid and Solid States (Fig. 81) The chief conditions for unlimited solubility in the solid state are that the two components should have the same type of crystal lattice and that the sizes of the atoms be very similar; the difference in size for iron-, nickel-, or cobalt-base alloys must not exceed 8 per cent. A difference in size over 15 per cent prevents the formation af solid solutions due to the extreme distortion of the solvent crystal lattice. If the atoms differ in size by more than 15 per cent, the formation of a chemical compound with a different crystal lattice or of a simple mechanical mixture of the components may be more probable from an energetical point of view.Systems with Mutually Soluble Components 109 The fact that elements have melting points near to each other and that they occupy neighbouring positions in the periodic table will facilitate the formation of a continuous series of solid solutions. As mentioned earlier, the crystal lattice of a metal consists of positively charged ions held together by uniformly distributed electrons which are free to move from atom to atom. The ratio of the number of valence electrons to the number of atoms is called the electron ratio. Each type of crystal lattice has its critical electron ratio at which it reaches its maximum stability. The critical Temperature CW 2 08 6) pempaent Ra wD E mae ‘% component A tae —— Fig. 81. Constructing an equilibrium diagram for a system of components that are completely mutually soluble in both the liquid and solid states electron ratio, determined on the basis of theoretical calculations for a face-centred cubic lattice is 1.36 and fora body-centred cubic lattice 1.48 electrons per atom. A continuous series of solid solutions may be formed only when the electron ratio remains constant or does not exceed the critical value at any composition of the components. Corjormation with electron ratio rule is an important condition for the formation of a continuous series of solid solutions in many systems. It must be noted, however, that such series are not always formed even when the electron ratio is suitable, This shows that solid solution formation is a complex chemical interaction whose laws are still insufficiently clear. Fig. 8! shows how an equilibrium diagram is drawn for components that are completely mutually soluble in both the liquid and solid states. In the complete form this diagram is illustrated in Fig. 82. The upper line corresponds to the temperatures at which the alloysuo Binary Alloys. Constitution and Equilibrium Diagrams begin to solidify, the liquidus. The lower line, the solidus, indicates the completion of solidification. In the temperature interval between the liquidus and solidus, the alloys are in a semi-solid state, i. e., they consist of crystals of a solid solution of components A and 8 and the liquid alloy. Solidification begins with nucleation and subsequent growth of the nuclei. Nucleation occurs in regions of the liquid phase where, Liguid phase Liguid phase 7] crystals of solid nA ty? solution {A+8) Cn) o 0 0 wm 0 6 0 DD component Bo Fig. 82. Equilibrium diagram for system of components that are completely mutually soluble in both the liquid and solid states due to fluctuations in concentration, the chemical composition corresponds to the equilibrium composition of the solid solution at the given temperature. The critical size of the nuclei depends upon the degree of supercooling and supersaturation of the initial liquid solution. The higher the degree of supercooling and supersaturation, the smaller the critical size of the nuclei. We shall now consider the solidification of an alloy containing 60 per cent B (Fig. 82). Freezing begins at temperature f, where the first crystals of the solid solution of the components A and B separate from the liquid alloy. The alloy is in a two-phase state below temperature /,. As indicated by the phase rule for a two-component system in which two phases are present (liquid and solid), the number_. Systems with Mutually Soluble Components ut of degrees of freedom will be F=2+1—2=t Therefore, each temperature corresponds to definite concentrations of the phases. The concentration of the solid solution, in equilibrium in the liquid phase, is determined by the point of intersection of a horizontal ine, passing through the given temperature, with the solidus (Fig. 82). The concentration of the liqui pase is determined by the intersection of the same temperature line with the liquidus. At a temperature of ¢,, for example, points m and n represent the concentrations of the solid and liquid phases, respectively. At temperature f,, they are represented by points m, and a,. Thus, in solidification, the composition of the liquid phase continuously varies along the liquidus while that of the solid phase varies along the solidus. The ratio by weight between the liquid and solid phases at a temperature f,, for example, may be calculated using the lever rule: weight of Hguid phase, em weight of solid phase fn An alloy containing 60 per cent B completely solidifies at temperature %. Fig. 83 illustrates the equilibrium diagram for copper-nickel allo: which are an example of complete mutual solubility in both the liquid and solid phases. In the solidification of alloy systems of this type, the solid solution crystals, separating during the process, are of variable composition. However, if the alloy is held at a sufficiently high temperature for ample time and is cooled very slowly, diffusion between the liguid part of the alloy and the crystals and also between the crystals of the solid solution, that are of varied composition, will help to establish equilibrium between the phases. Ditfusion also facilt- tates equalisation of the composition after the alloy is compietely solid. Since diffusion is a very slow process, the composition does not have enough time, in actual practice, to be fully equalised and the crystals usually vary in composition. The axes of the crystals, which start to grow at the beginning of solidification, are usually richer in the component having the higher melting point. The peri- pheral layers of the crystals and the spaces between the axes are filled in later and will be enriched by the component lowering the melting point of the alloy. No true equilibrium is attained, for such cases, between the liquid and solid phases in solidification. Therefore, the diagram, which is drawn for equilibrium cases, cannot be used for determining compositions of the phases by intersections on the liquidus and solidus.112 Binary Alloys. Constitution and Equilibrium Diagrams Upon faster cooling, the structure of solid-solution type alloys will be of a clearly defined dendritic character (Fig. 84) due to the liguid phase Temperaters, C—om Satta solution {0u-Wi] a 0 20 30 40 50 50 70 6a 90 100 % nicked — Fig. 83. The copper-nickel equilibrium dtagram Fig. 84. Microstructure of a solid sotution of 90 per cent Cut 10 per cent Zn: a—dendrites of the solid solutfon (as-cast), x 3, and 6-—polyhedrons ot the solid solution (after homogenising or diffusion annealing), x 100 lack of homogeneity in the composition of the crystals. Such inho- mogeneity of the alloy within the crystals is called dendritic segregation or coring.Systems with Mutually Soluble Components 3 The larger th: temperature interval between the liquidus and solidus, the greater the degree of dendritic segregation. It may be reduced somewhat by holding the alloy for a long period at a temperature high enough toensure a suitable rale of diffusion. After such treatment (annealing), the dendritic structure of the cast alloy is no ionger evident and polyhedral grain structure of the homogeneous solid solution may be observed (Fig. 84, 6). The variation in mechanical, physical and fabrication properties of alloys that form a continuous series of solid solutions is shown in Fig. 85. The diagrams indicate that solid solutions possess higher strength and hardness than their components. In many cases, however, this increased tensile strength is not accompanied by reduced ductility. On the contrary, ductility is often increased together with tensile strength and hardness (for example, in & brasses). N. S. Kurnakov first noted that the formation of a solid solution always increases the electrical resistivity. This is associated with the considerable distortion of the electrical field of the solvent metal by the atoms of the solute metal. It must be noted, however, that the electrical resistivity of a solid solution remains practically unchanged upon an increese in temperature while that of pure metais substantially increases. Therefo- re, at high temperatures, the resistivity pure ntetais may exceed that of solid tions.* Aileys which form homogeneous sslid solutions are ordinarily easily worked (by rolling, forging, stainping, etc.) but have comparatively low casting properties (they have a tendency to form cracks ana dispersed porosity). Such alloys are machined with difficulty in many cases duc to their high u- * This intlicates that the telationsh!e deduced by N.S. Kurnakoy is valid only at suflicieitly low temperatures. Solid solution At Uitimate strength 6,, kgsan? Temperature a i & 2 § ‘ 3 8 3 g 8 3 3 Toval chaage t voiume i Fone povastig %pl hirinkrge ose 4 60 60 WP Pe Fig. 85. Reinticnship between Prupertic: and com- positica of alleys that form . conuuy us series of salid solations14 Binary Alloys. Constitution and Equilibrium Diagrams ductility. Substances which reduce the ductility are added to these alloys to improve their machinability. Alloys forming homogeneous solid solutions are widely used as engineering materials. b. Equilibrium Diagrams of Systems Whose Components Have Complete Mutua! Solubility In the Lista Sine and Limited Solubility in the Solid State and in Which the Solid State Solubility Decreases with the Temperature Fig. 86, a shows how an equilibrium diagram of this type Is drawn while Fig. 86, 6 is the complete diagram on which the obtained structures are indicated. The lines ae and be are the liquidus. Crystals of a solid solution of component B in component A («) begin to precipitate from the liquid alloy along line ae; the solid solution of component A in component B (p) precipitates along line be. The lines ad, dec, and be correspond to the solidus. The point d corresponds to maximum solubility of component B in component A at the eutectic temperature (#,). Point ¢ is the same for component A in component B. : In a similar manner, points & and f irfdicate the solubilities of component A in component B, and B in A, Tespectively, at normal room temperatures. The lines d& and fc thus show the variation in solubility with the temperature. At point e, which conforms to the conditions of nonvariant equilibrium (F=2+-1—3=0), the solid solutions « and B simultaneously separate from the liquid phase to form the eutectic (@-+B). Let us consider now the solidification process of alloys 1, 2, and 3 (Fig. 86, 0). Alloy | begins to solidify at temperature f, and finishes at ¢,. The solid alloy contains only crystals of the'solid solution a and no further phase transformations occur at lower temperatures. Alloy 2 begins to solidify at temperature ¢, and finishes at #,. The composition of the liquid phase varies during solidification along the liquidus; that of the solid phase varies along the solidus. Thus, at temperature ¢, points m and n determine the compositions of ‘ne liquid phase and the solid solution crystals, respectively. Wren it is completely solid, alloy 2 will consist only of a solid solution crystals. Upon further cooling to temperature 4,, the solid solution will become supersaturated and, at lower temperatures, it will decompose and the surplus component B will be separated as B ecystals whose amount increases as the temperature falls. Therefore, at temperatures below #,, the alloy will consist of two phases a+f (Fig. 87). The composition of the a and B solutions varies along line det @ crystals and along ef for 6 crystals upon a fall in temperature.GOLA «RRR Temperature, C-—— tim me ‘compat ome 200 07 8 4 | Keompment A Temperature —»— Fig. 86. Equilibrium diagram of a system of components that have complete mutual solubilily in the liquid state and limited solubility in the solid state. @—consiructing the diagram, 6—the finished diagram116 Binary Alloys. Constitution and Equilibrium Diagrams For example, at temperature ¢’, the composition of the a-phase is determined by point 6 and that of the B-phase, by point y. The quantitative relation between the weights of thea- and 6-phases conforms to the lever rule: amount of a-phase _ 2’) amount of B-phase 7° On the basis of energetic and concentration fluctuations, nuclei of the B-phase are first formed on the boundaries of the original a-phase m Fig. 87. Separation of the surpius phase upon decomposition of the solid solution (AI-Cu alloy), X 250 grains (Fig. 87). Rapid cooling (quenching) will suppress the formation of B-phase crystals and a homogeneous alloy may be supercooled io low temperatures. A quenched solid solution of this type, supersaturated with component B, is unstable and will decompose upon heating and, for some alloys, even at norma} temperaiure. The rate of decomposition increases with the temperature to which the alloy is heated, This separation of the surplus phase due to decumposition of a supercooled solid sotution is accompanied by considerable changes in the properties of the alloys. An increase ii hardness is characteristic for this process which is called age hardening or precipitation hardening. Alloy 3 (Fig. 86, 6) begins lo solidify at temperature f, and is completely solidified at temperature ¢, In solidification, the composition of the liquid part of the alloy varies continuousiy along the liquidus, approaching the eutectic composition (point e); the composition of the solid phase varies along the solidus moving toward the maximum solubitity (point 4). The telation between the liquid and solid phases may be found at any temperature by the lever rule. At temperature ¢,, the liquid phaseSystems with Mutually Soluble Components u7 teaches the eutectic composition and the alloy completely solidifies. In this, alpha and beta solid solutions, forming the grains of the eute- clic, are simultaneously precipitated from the liquid phase. The solidified alloy will consist of primary fully saturated (point d) crystals of alpha solid solution, precipitated in the temperature interval 1,-¢,, and the eutectic, formed at temperature ¢, and comprising alpha and beta solid solutions of the extreme compositions (points d and c, respectively). Thea crystals, both primary and those Fig. 88. The lead-tin equilibrium diagram in the eutectic, decompose upon a further drop in temperature due to the decrease in solubility. As a result beta solid solution crystals precipitate from the a crystals. The compositions of the a-phase and the B crystals, both surplus and those in the eutectic, vary with the temperature along the lines dk and c/, respectively. After the complete jolleitication of an alloy, their compositions are indicated by points & and f. . An alloy of the composition corresponding to point e (60 per cent B) will contain only grains of the eutectic (¢+) after solidifying. Hypereutectic alloys will solidify in exactly the same manner as the hypoeutectic ones considered previously except that primary 6 crystals will precipitate instead of primary a crystals. Equilibrium diagrams of this type are of great practical significance since the properties of the alloys (for example, alloy 2) can be substantially changed by heat treatment (hardening and ageing). The Pb-Sn system has an equilibrium diagram of the above-described type; this diagram is shown in Fig. 88. N. S. Kurnakov has shown that the properties of -alloys of this type vary along a curve in the region of homogeneous solid solutions and along a straight line for regions of two-phase mixtures (Fig. 89).us Binary Alloys. Constitution and Equilibrium Diagrams Tendency to form et crdohs porasity Fig. 89. Relationship between properties and composition in a system with imited solubility in the solid state In the second case, the end points of the straight line correspond to the properties of fully saturated solid solutions. It must be noted that the properties, in the regions in which two phases (@ and B) exist, depend upon the degree of dispersity of the mixture (Fig. 89). The properties of highly dispersed mixtures wil! not conform to the straight line relationship. Investigations conducted by A. A. Boch- var and co-workers have found a definite relationship between the composition of an alloy and its casting properties (Fig. 89). The larger the temperature interval of solidification, the less the fluidity of an alloy. The greater the distance between the liquidus and solidus (in a horisontal direction), the more tendency an alloy has for dendritic segregation. Fluidity is noticeably increased and the tendency to form hot cracks is reduced in alloys which form a eutectic in solidifying. Shrinkage appears as a concentrated pipe in alloys which solidify in a narrow freezing range and, particularly, in eutectic alloys which solidify at a constant temperature. On the other hand, shrinkage in alloys, solidifying in a wide freezing range, is found chiefly in the form of dispersed porosity (Fig. 89) throughout the casting. This may necessitate rejection of the casting. Therefore, fo ensure good casting properties, the concentration of the components must exceed tneir maximum solubility in the solid state and should approach the eutectic concentration. Alloys in the region of homogeneous so- Hd solutions possess high ductility and, therefore, are easily rolled, forged, pressed, etc. Plastic deformation of alloys consisting of a mixture of several phases is much more difficult. Ductility is reduced tothe greatest extent when the eutectic appears in the structure, Consequently, the maximum solubility at the eutectic temperature is the upper limit of concentration for wrought alloys.Systems @ith Mutually Soluble Gomponents 119 Alloys of the type here described are extensively used in engineering. c. Equilibrium Diagrams of Alloys Whose Components Have Complete Mutual Solubility in the Liquid State and Limited Solubility In the Solid State (Alloys with a Peritectic Transformation) The equilibrium diagram for this type of alloy is shown in Fig. 90. This diagram differs from the preceding type (Fig. 86) in that the crystals of beta solid solution, precipitated at the beginning of solidification, react with the liquid alloy of a definite composition to Liguid phase o 0 0 WH 4 55 8 RM & WM WO Yecompanent B Fig. 90. Equilibrium diagram of an alloy subject to peritectic transformation form new crystals of alpha solid solution. This transformation or reaction occurs at a constant temperature (as do eutectic transformations) and is called perttectic transformation. In the equilibrium diagram (Fig. 90), the line abe is the liquidus and adec is the solidus. Point d represents the maximum solubility of component B in component A at temperature ¢,; point eis the same ior component A in component B. Points & and f represent maximum solubility at normal temperatures. Thus, lines dk and ef show the120 Binary Alloys. Constitution and Equilibrium Diagrams variation in solubility in the alpha and beta solid solutions upon cooling. The diagram can be better explained by considering the solidification of alloys /, 2, and 3 which contain 40 per cent B and 60 per cent A, 61 per cent B and 39 per cent A, and 80 per cent B and 20 per cent A, respectively. The 80 per cent B alloy begins to solidify at temperature ¢, when crystals of beta solid solution precipitate from the liquid alloy. At temperature ¢, (Fig. 90), the liquid phase has the composition conforming to point b and the crystals of solid solution are enriched by component B to the maximum concentration, shown by joint e. Here the alloy completely so- idifies according to a peritectic transformation that consists in an interaction between the liquid alloy of composition 6 with the fully saturated B crystals (point e). This forms a new alpha solid solution of the maximum concentration (point d): Ly thy Gat Be For any given composition of the 1g, 91. F xe phase struc - : ; indi sea result of peritectic tae alloy, the proportion of the reacting nsformation (Cu-Zn alloy), x 250 phases is characterised by the amount of excess B crystals and, therefore, a certain amount of f crystals remains unused in the peritectic transformation. ‘The structure cf the solid afloy will be a peritectic mixture of two solid solutions («+B) whose compositions will vary along the lines di and ef at a further fall in temperature. Fig. 91 illustrates the structure of an actual alloy after a peri- teciic transformation. The light crystals are a while the dark ones are B. of peritectic reaction, in which three phases participate (liquid and crystals of alpha and beta solid solutions), corresponds to nonvariant equilibrium, i. e., the process proceeds at constant temperature: F=24+1—3=0 Alloy 2, containing 6] per cent B, begins to solidify at temperature ¢, and finishes at the peritectic temperature ¢,. The alloy completely solidifies at a constant temperature. This is accompanied by the peritectic transformation: LytB, + mgSystems in Which Chemical Compounds are Formed 121 Thus, the result of the peritectic transformation (for the given alloy) will be the disappearance of both initial phases (liquid phase of composition 6 and crystals of the beta solid solution of composition e) and the formation of a single alpha solid solution. At temperatures below #,, however, the alpha solid solution will be partially decomposed, resulting in the precipitation of surplus beta solid solution crystals. The compositions of the alpha and beta solid solutions will vary along the lines de and ef which indicate the maximum solubility at various temperatures. After being cooled to room temperature, the alloy will consist of separated crystgis of alpha and beta solid solutions, having the compositions & and f. Alloy /, containing 40 per cent .B, begins to solidify at temperature ft, and crystals of beta solid solution precipitate from the liquid phase. The following peritectic transformation occurs when temperature f, is reached: Ly + Be Ly tae The liquid phase is in excess after the peritectic transformation in this alloy. Thus, the solidification of this alloy is not completed at the temperature of peritectic transformation and crystals of alpha solid solution continue to precipitate from the liquid alloy upon jurther fall in temperature. The alloy completely solidifies at temperature f, and consists only of crystals of the alpha solid solution. It is evident from Fig. 90 that alloys having a component B concentration less than 47 per cent, i. e., located to the left of point &, cnist solely of alpha solid solution crystals. Alloys containing more than 96 per cent component B (to the right of point f) consist only of beta solid solution crystals. Peritectic transformations occur in a sumber of alloys which are important engineering materials. They include: Fe-C (Fig. 108), Cu-Zn (Fig. 257), Cu-Sn (Fig. 260), and others. 6-5. Equilibrium Diagram of a System in Which the Components Have Unlimited Solubility in the Liquid State and Form Chemical Compounds Upon Solidification One or several chemical compounds may be formed in many metallic systems. Chemical compounds are classified in the following three main types: 1) valency compounds, 2) electron phases, and 3) interstitial phases. Valency compounds include MgsSi, Mg.Sn, MgsPb and other compounds of magnesium with elements of the fourth, fifth, and sixth groups in the periodic system. A characteristic feature of these compounds is the practically complete insolubility of the components.I Binary Alloys. Constitution and Equilibrium Diagrams Valency compounds are seldom formed in metal alloys. In the majority of cases, the components combine in intermetallic phases (chemical compounds) which do not follow simple valence rules. Electron compounds (Hume-Rothery phases) are of prime importance in metallic systems. Compounds of this type have a definite ratio of the number of valence electrons to the number of atoms, i. e., a definite electron ratio. Thus, a whole series of compounds exist in which the ratio is °/,; in others it equals */,, and */,. Each of these ratios corresponds to a definite type of crystal lattice. All compounds with anelectron ratio of */, have a body-centred crystal lattice. This type of compound includes CuZn *, Cu,Al, Cu,Sn, NiAl, FeAl, and others. They are known as the B-phases. Compounds Cu,Zn,, Cu,Al,, and others, with an electron ratio of "/,,, have ‘a complex cubic lattice. They are called the y-phase. Finally, such compounds as CuZn,, CusSn, Cu,Si, and others with a ratio of 7/,, have a close-packed hexagonal lattice and are ~ catled the e-phase. A large group of metals (Fe, Cr, Mo, W, Ti, V, and others) form chemical compounds (carbides, nitrides, etc.) with carbon, nitrogen, hydrogen and boron, all of which have a small atomic radius. These az called interstitial compounds. This group of compounds may be formed if the ratio of the atomic radius r, of the nonmetal (C, N, H, or B) to the atomic radius 7, of the metal is less than or equal to 0.59 (= <0.59). Interstitial sompounds have the following types of chemical for- mulae: Me,X [Mn,N, Fe,N, etc.]., Me,X([W,N, Fe,N, etc.], and’ MeX (TiN, TiC, WC, VC, etc.]. The atoms of interstitial compounds form one of the most typical cubic or hexagonal crystal lattices into which the atoms of the me- lalloid are inserted, occupying definite pores or vacant positions. Interstitial compounds or phases differ from interstitial solid solutions in that their crystal lattices differ from those of the metals of which they are formed. They have certain typically metallic properties (electrical conductivity which is reduced with an increase in temperature). They also have an exceptionally high hardness and melting point. Solid solutions, usually with excess metal atoms, are easily formed on the basis of interstitial compounds. a. Equilibrium Diagram of a System Whose Components Have Complete Mutual Solubility In the Liquid State and Form Stable Chemical Compounds After Solidification In this diagram, the components A and B form a chemical compound A, 8, which has a constant melting point, shown on the diagram by * In compounds copper provides one valence electron, zine provides two, aluminium three, tin four, while nickel and iron do not provide any.sks oy sway tre Pato} st punoduios jeauaqs [quis 9d-BW pum ‘ug-3W ‘Ig-3W 2y3 205 sUIELBeP oMqunbg, “6 “My GPA UF widIshs jo WeABeIp wnnqyinby -7 By a % ew St neny % Rwoy a 2 1 Ry] = z| © J, ‘singasadusas124 Binary Alloys. Constitution and Equilibrium Diagrams point ¢ (Fig. 92). Under a microscope, a chemical compound is observed as separate polyhedral grains, the same as in a pure metal. The effect of the existence of a stable chemical compound on the equilibrium diagram is that en ordinate, corresponding io the compo- Temperature, Cm Hardness, Boh Electrical resistivity Electrical conductivity 0 2040 60 80 HO % component 8 Fig. 94, Relationship. between properties and com. position in a system in which a stable chemical compound is formed sition of this alloy, divides the diagram into two parts (Fig. 92). We obtain what may be considered as two diagrams of the first type, A-A,B, and A,B,-B, joined along the ordinate corresponding to the chemical compound composition. Alloys A-A,B,, begin to solidify at the temperatures indicated by the line abc. A crystals begin to separate on linead and crystals of the compound A,B, along line bc. lidification is completed at the eutectic temperature (line bf). A, 8,,-B alloys begin to solidify along line ede. On line ed crystals of compound 4,8, separate and B crystals along line de. In this case, solidification is completed along the eutectic line ndm. It is evident from the diagram (Fig. 92) that two eutectics are formed. One, represented by point 4, consists of a mixture of A crystals and compound A,8,; the second, tepresented by point d, consists of compound A,B, and B crystals. In the first part of the system (4-4,8,,), hypoeutectic alloys begin to solidify with the precipitation of primary A crystals from the liquid alloy; crystals of the compound A,B, precipitate in hypereutectic alloys. In the second part of the system (A,,B,,-B), hypoeutectic alloys contain primary crystals of compound A,B, in addition to the eutectic (A,B,+B); hypereutectic alloys contain Berystals. Fig. 93 illustrates equilibrium diagrams of three actual systems in which stable chemical compounds are formed. It must be noted, however, that in most cases, the components and the chemical compound form limited regions of solid solutions. In these cases, the diagram will be complex and will consist of two diagrams of the eutectic type having limited solubility in the solid state (Fig. 94). The y region is a solid solution on the basis of the chemical compound or a compound of variable composition. Since the typical equilibrium diagram given124 Binary Alloys. Constitution and Equitibriam Diagrams point c (Fig. 92). Under a microscope, a chemical compound is observed as separate polyhedral grains, the same as in a pure metal. The effect of the existence of a stable chemical compound on the equilibrium diagram js that an ordinate, corresponding to the composition of this alloy, divides the diagram into two parts (Fig. 92). We obtain what may be considered as two diagrams of the first type, A-A,B,, and A,8,-8, joined along the ordinate corresponding to the chemical compound composition, Alloys A-A,B, begin to solidify at the temperatures indicated by the line abc. A crystals begin to separate on line ab and crystals of the compound A,8,, along line bc. Solidification is completed at the eutectic temperature (line Rbf). . A, B,-B alloys begin to solidify along line ede. On line cd crystals of compound A,B, separate and B crystals along line de. In this case, solidification is completed along the eutectic line ndm. It is evident from the diagram (Fig. 92) that twoeutectics are formed. One, represented by point 6, consists of a mixture of A crystals and compound A,8,; the second, represented by point d, consists of compound A,B, and B crystals. In the first part of the system (A-A,8,), hypoeutectic alloys begin to solidify with the precipitation of: primary A crystals from the liquid alloy; crystals of the compound A,B, precipitate in hypereutectic alloys. In the second part of the system (A,8,,-B), hypoeutectic ai loys contain primary crystals Fig. 94. Relationship bet- ©! compound A,B,, in addition to the eutec- ween properties ane com- tic (A,B,+-B); ‘hypereutectic alloys contain position in a system in B crystals. Fig. 93 illustrates equilibrium which @ Stable | chemical diagrams of three actual systems in which compound is ton stable chemical compounds are formed. It must be noted, however, that in most cases, the components and the chemical compound form limited regions of solid solutions. In these cases, the diagram will be complex and will consist of two diagrams of the eutectic type having limited solubility in the solid state (Fig. 94). The y region is a solid solution on the _basis-of thé chemical compound or a compound of variable composition. Since the typical equilibrium diagram given Wardness, Bh Electritat resistivitySystems in Which Chemical Compounds are Formed 125 in Fig. 94 is identical to a diagram previously explained (Fig. 86), there is no necessity for following the solidification of various alloys in this system. Fig. 94 also shows the relation between properties and composition for alloys of this type. Characteristic features of chemical compounds are their exceptional hardness, high electrical resistivity, and unsu- itability for plastic deformation; they are very brittle. On the properties vs composition curves, there is a sharp bend in the straight line at compositions corresponding to the chemical compound. The point, at which the curve bends sharply, corresponds to the composition of the chemical compound and is called the singular point. The singular point helps to exactly locate the position of the chemical compound on the equilibrium diagram. b. fli brim of a System of its that Form an Unstable Fecal ponds ilditication Ned Which Decomposes soon Chem! Reheath Before Melting, Peritectic Reaction, Form! Soll ee Siete ate a Liquid Phase ing « Fig. 95 illustrates the diagram obtained in this case. The chemical compound A,B, is located at the ordinate 80 per cent 8. This compound ty Liguid phase Liguid phase Temperature 109M) 4x} 60D 0 0 5D 6 1M 60 9 %canpaneat 8 : Fig. 95, Equilibrium disgram of a system in which an unstable chemica} compound ® * = 1s formed pe is formed at temperature f, as a result of a peritectic transformation of the liquid phase, having a composition indicated by point d (L,),126 Binary Alloys. Constitution and Equitibrium Diagrams and the beta solid solution of maximum concentration (point e): LatB, ~ AnBy A chemical compound is formed only in alloys of the indicated composition. Alloys of higher concentration (located to the right of 80% B) will contain excess crystals of B solution, in addition to the compound A,B,; those of lower concentration (to the left of 80% B) will contain the eutectic (2+A,B,q)- 6-6. Equilibrium Diagram of a System Whose Components are Subject to Allotropic Transformations Phase transformations occur in the solid state of systems in which ~ one or both components are subject to allotropic transformations, for example, in systems of which Fe, Co, Mn, Ti and Zr are components. Fig. 96 shows an equilibrium diag- tam for the case when the components of the system solidify as pure metals and the component A has two allotropic forms, « and B. The transformation of component A from allotropic form @ to form B s, takes place at a constant temperature ¢,, since three phases a-A, B-A and B are in equilibrium during the. transformation (F=2+1—3=0). A 4 characteristic case of this type of diagram is the Fe-S system (Fig. 3 Liguid phase Liquid phase a) Temperature, °C, If alloys, in solidifying, form so- 0 2 40 6 & 1 \id solutions of components that ‘A companent 8 may exist in different allotropic Fig. 96. Equilibrium diagram of a forms, various relationships may system in which one of thecompone- occur between them. vats has two allotropic forms The phase diagram of a case, in which the high-temperature modifications have complete mutual solubility and the low temperature modifications have but limited solubility, is illustrated in Fig. 97. Solid state transformations may be described in the same way ay in solidification from a liquid solution * (see page 114). It is evident that the line ac corresponds to the beginning of decomposition of the beta solid solution with the precipitation of «’ solid solution * In real systems, solid state transformations take place in a more complex manner.Diagrams of Ternary Systems 127 crystals. On line cb, «” solution will precipitate. The beta solid solution will decompose into a mixture of the a’- and a”-phases along line dee at a constant temperature (F=0). The process is similar to a eutectic transformation except that the initial phase is a solid solution. This is called a eutectoid trans. formation and the mixture of crystals (a’-++«”} is called the eutectoid. liguid phase Liquid phase + Jemperature, C id | a , + + leutectoit eutectoid ectt fatvat) | (a'va*} | | 00 2 30 40 6 7% % 90 100 % component 6 Fig. 97, Equilibrium diagram of a system in which the high-temperature allotropic forms of the components have complete solubility, and the low-temperature forms— limited solubility It should be noted that solid solutions differ from liquid solutions in that they have a larger tendency to supercool and produce nonequilibrium states which, naturally, cannot be shown on the equilibrium diagram where all phases are in equilibrium. 6-7. Diagrams of Ternary Systems Since ternary and more complex alloys are widely used in engineering, it is necessary to study the equilibrium diagrams for multiple-component systems. Such diagrams enable more expedient alloy compositions to be selected and help us to understand the va-128, Binary Alloys, Constitution and Equilibrium Diagrams rious processes that occur in such systems. The Russian school of hysical metailurgists, under the leadership of Academician N. S. Rummakov, have done extensive research in this field. Many ternary diagrams have been constructed and they are widely used in actual practice. The diagram of a ternary system is a three- dimensional model constructed on a base which is an equilateral triangle. The concentration of the components may be shown in the triangle in the same scale. This is called a concentration triangle. a B £L a g ie 4 i] oooA 7 o é& c Fig. 98. Determining the composition of ternary alloys The components of the alloy are located at the corners of the triangle while the sides represent the three binary alloys. Any point within the triangle specifies the composition of a ternary alloy. A property of an equilateral triangle is frequently used to determine the composition of an alloy within the triangle. This geomet- tical rule states that the sum of three lines (on-+om--op) (Fig. 98, a), drawn parallel to the three sides, from any point within the triangle, is equal in length to one side ¢ of the triangle which is iaken as 100 per cent. To find the composition of the alloy at point o, we simply draw three fines from this point parallei to the sides of the triangle. The corresponding lengths of the Jines show the percentages oi components A, B, and C in the alloy. The composition may also be determined by snotier method. In this case, the altitude of the triangle is taken 2s 106 per cent and the geometrical theorem which is used state: that the sum of ihree lines drawn perpendicular to the three sides from a point in an equilateral triangle, is a constant equal to the altitude (8B ,=ima+ -+-me-+mb) (Fig. 98, 6). On the basis of the properties of equilateral triansies, it is nat difficult to show that: 1) All alloys whose compositicn lies on a straight line connecting an apex of the triangle with the opposite side have 2 constant composition ratio of two components, for exempie, aliovs on the tine Aa have a constant ratio of componenis B and C (Fig. 98, ¢).Diagrams of Ternary Systems tag 2) All alloys which lie on altitudes of the triangle (Fig. 98, c) have the same compssition for two of the components, for example, alloys on the altitude Bd contain equal amounts of components A and C. 3) All alloys which Ite on a line parallel with one side of the triangle (ed) have the same content of the component located at the apex opposite which this line is drawn (B in Fig. 98, c). The composition of ternary alloys is represented on a plane while a three-dimensional model is required to show temperature transformations. In constructing a ternary diagram, the first step is the same as for binary alloys where cooling curves are plotted in temperature vs time coordinates for alloys of various composition. When this has been done, the points representing the various alloys are plotted in the composition triangle. Perpendiculars are constructed to the plane of the triangle at the corresponding points and the temperatures of the critical points for each alloy are located on the cor- fesponding perpendicular. Surfaces passing through the critical points, in the same way as lines were drawn through the critical points in binary diagrams, will provide a three-dimensional repre- sentation of the ternary diagram. Ternary diagrams are classified in accordance with the same features as for binary alloys, i.e., on the basis of the solubility of the components in the fiquid and solid states, presence of chemica! compounds, etc. Fig. 99 illustrates a ternary diagram for a system in which the com- jonents have complete mutual solubility in the liquid siate and orm a simple mechanical mixture when they solidify. Binary systems A-B, B-C, and A-C are constructed on each side of the composition triangie (Fig. 99). The surface 4,6, E’E; ¢, is a part of the liguidus and corresponds to temperatures at which component A begins to precipitate [rom the liquid alioy, surface t,£, E’E; t, is the same for compouent 8 and surface #,£, EE; ty refers to precipitation of component C. The lines Ey E’, EE’, and & &’ represent the binary eutectics in the ternary alloy. All three surfaces composing the solidus intersect at the common point &’ which is the ternary eutectic. The projeciions of the fines of the binary of the composition triangle give three two-di: F,£, and E,E (Fig. 99, 5). We may follow the solidification of an alloy having s composition represenied by point @ to better understand this disgram. It is evident from Fig. 09 that at point R the alloy is in a Equid state, When the temperature 1s lowered to point /, the liquid alloy de- comes saturated with component A which begins to precipiizte at this point, At this, the liquid alloy will become richer in co::.panents B ectics ov the plane ional carves £,8,130 Binary Alloys. Constitution and Equilibrium Diagrams and C and its composition will change along the surface of the fiqui- dus in the direction 10’. The projection of the curve 16’ on the composition triangle will bea straight line ab passing through the apex A *. When the composition of the liquid alloy reaches point b, it wil] be saturated with Tr VR a Fig. 99. Ternary diagram (model) for a system of components having complete mutual solubility in the liquid state and forming a mechanical mixt- e ure in the solid state components A and B. Therefore, both of these components will solidify at a further drop in temperature and form a binary eutectic. This binary eutectic (A+) is formed along a temperature interval (F=3+1--3=1). In this temperature interval (formation of the cutectic), the composition of the liquid alloy will vary along line £; E', from point 6’ to the ternary eutectic E’. In the projection on the composition triangle, this change will follow along curve £,£, from point 6 to point £ (Fig. 99, 6). * Though the composition of the liquid ailoy varies, the proportions of the components B and C remain constant since their absolute amount in the alloy does not change. This requires the composition of the liquid alloy to vary along a straight line passing through an apex of the triangle.Diagrams of Ternary Systems 131 When the liquid alloy reaches the concentration represented by point £(E'), the ternary eutectic (A+B+C) is formed. This takes place at a constant temperature (F=3+1—4=0) corresponding to point 2 (Fig. 99) in the eutectic plane (A’B’C’) passing through point £’. The completely solidified alloy will consist of A crystals, binary eutectic A+B, and the iL ternary eutectic A+B+ 4b +C. Another alloy with the composition represented by point m will solidify in exactly the same manner. After solidification this alloy (alloy //} will consist of B+(B+C)+ (A+B+C)}. The latter 4, c Fig. 100. Ternary diagram Fig. 101. Ternary diagram (model) for a sys- (model) for a system of com- tem of components icving limited solubility ponents having complete mu- In the solid state tual solubility in both the Ii- quid and solid states two phases are the binary and ternary -stectics, respectively. Alloys located on the lines £,£ and ££. tor example, alloy Mi in Fig. 99, will consist, after solidifying, of “se binary and ternary eutectics. Alloys on the lines AZ, BE, and Cé will contain no bina- ty eutectics; they will consist only of an ex.ess component (A, B or C) and the ternary eutectic (A+B+C). The alloy at point £ will consist only of the Ternary eutectic (A+B+0)}. A ternary diagram for components having complete mutual solubility in both the liquid and solid states is shown in Fig. 190. This132 Binary Alloys. Constitution and Equilibrium Diagrams diagram has two surfaces, liquidus and solidus, between which the ternary solid solution is solidified. The solid alloy consists of this ternary solid solution. oo Fig. 101 illustrates a ternary diagram of a system having limited solubility in the solid state. ; Quite often, only certain known sections of a ternary diagram are used to show transformations. There are several methods for constructing such sections: Vw A vy N Fig. 102. Isothermic sections of a ternary diagram } Horizontal sections of the ternary diagram are made, either in the ferm of isothermal sections, which represent the phase and siruciural composition of all the alloys at a definite temperature, tions of certain surfaces and lines on a horizontal sur- miposition triangle). al sections are made either parallei with one side of the triangle or passing through one of its apexes. Some- al sections are constructed. These sections show the phase and siccciural composition for all temperatures but only for definite compositions of ternary alloys. ed most frequently is a vertical section of a ternary a constant composition of one component or with constant proportions between two of the components. Pig. 12 dluctrates isothermic sections of a ternary diagram (Fig: 99) which are horizontal planes made at a temperature above the temperatures for ihe pinary and ternary eutectics (a) and at a temperature slightly svove the ternary eutectic point (6). Fig. 102, ¢ shows a horisontal section below the ternary eutectic point for alloys forming limited solid solutions (Fig. 101). Several vertical sections in planes parallel with side AB are shown in Fig. 105. In fhe section ab, points a, K, M, N, and ® are intersections of the plane in question with corresponding volumes of the ternary diagram (Fig. 99). In our section, the two surfaces of the liquidusDiagrams of Ternary Systems . a Section a,b, Section a,% Fig. 103, Vertical sections of a ternary diagram134 Binary Alloys. Constitution and Equilibrium Diagrams intersect in point M. The intersection with the surface of the liquidus is represented by the lines a” M and Mo’’. Further down, the vertical section cuts the regions in which the binary eutectics precipitate from the liquid alloys. This is represented by the lines a’ X and Nb". Points XK and N are on the surface of the solidus, i.e., at the temperature of solidification of the ternary eutectic. They can therefore be connected by a straight line parallel with the base. Sections a1b,, Gsbs, and aybs were obtained in a similar manner.Chapter 7 THE IRON-CARBON EQUILIBRIUM DIAGRAM Afloys of the iron-carbon system include steel and cast iron, which are of the most vital importance to modern industry due to their extensive, versatile applications. Alloys with a carbon content up to 2.0 per cent are called steels; those with a carbon content exceeding 2.0 per cent are called casi trons.* Earliest research on the iron-carbon equilibrium diagram was carried out by the great Russian scientist D. K. Chernov.** He first showed, in 1866, that when steel is heated (or cooled), transformations, associated with structural changes, occur at definite temperatures (depending on the composition of the steel). Chernov’ called these temperatures “peculiar points” (critical points) and designated them with the letters a and 6. At the present time Chernov’s points a and 6 are designated as A, and A,, respectively. Chernov's discovery of critical points in steel has served as a basis for the development of the science of metals. The works of Academicians A. A. Baikov, I. V. Gutovsky, and N. P. Chizhevsky, as well as A. A. Rzheshotarsky were a great con- tribution to the study of the iron-carbon system. The investigations of such famous foreign scientists as H. C. Sor- by, F. Osmond, W. Roberts-Austen, and P. Goerens helped to construct the iron-carbon equilibrium diagram as we know it today. 7-1. Iron The melting point of iron is 1,539°C. There are two allotropic forms of iron: @- and y-iron. Curves showing the variation in free energy with the temperature for a- and y-iron are given in Fig. 104. The a-iron free-energy curve * This classification of steel and cast iron, though accepted, is not exact since there are alloy steels containing more than 2.0 per cent C and alloyed cast irons containing less than 2.0 per cent C. ** D. K. Chernov, “A Critical Review of the Articles by Messrs Lavrov and Kalakutsky on Steel and Steel Gun Barrels and Research Made by the Author on the Same Subject", Transactions of the Russian Engineering Society, 1868.136 The Iron-Carbon Equilibrium Diagram has a gradually changing inclination. That of the y-iron curve increases shar ly at 910°C and Intersects the a-iron curve a second time at Moo Therefore, lng exists at temperatures up to 910°C and in the interval from 1,400° to 1,539° C; y-iron is stable at temperatures from 910° to 1,400° C. In the interval from 1,401° to 1, C, a-iron is often called 8-Iron. Fig. 104. Free-energy vs temperature curves for a- and. y-fron The cooling curve for pure iron is illustrated in Fig. 105. The crystal lattice of a-iron is of the body centred cubic type with a lattice constant of 2.86 A. Up to 768°C, a-iron is ferromagnetic. The maximum solubility of ‘carbon ina-iron is 0. .025 per cent (at 723°), at 20°C, the sotubility if only 0.0025 per cent.* The solid solution of carbon in a-iton is called ferrite. Under a microscope ferrite is seen as homogeneous polyhedral grains (Fig. 106). The mechanical properties of ferrite (0.06 per cent C) are characterised by the following values: o,=25 kg/mm"; — o,= 12 kg/mm!; “$50 per cents $= 80 per cent. has a face-centred cubic crystal lattice with a content ‘of 9) ena fat 910° C). It is very weakly magnetic (param: netic). The cerbon.solubility of y-iron at 1,130° reaches 2.0 per cent. * According to more exact data which have not been, entered in the diag- Tam, the ata solubility at 20°C Is about 10-* per cent.tron ‘ 137 The solid solution (interstitial type) of carbon in y-iron is called austenite, Fig. 107, a illustrates the crystal lattice of austenite. The carbon atom occupies the centre of the face-centred cubic lattice of y-lron. This carbon is in the form of ions which have been Temperatars, "C 8 8 3 8 § 8 8 8 §€ § § 8 Time Fig. 105. The cooling curve for pure tron fonised twice (raving given up two electrons for common use) while the iron stom has been ionised but once (having given up one electron). The microstructure of austenite Is shown in Fig. 106, 6. The structure of austenite was first observed by A. A. Baikov. In addition to solid solutions (ferrite and austenite), iron and carbon form a chemical compound, iron carbide Fe,C which has been named cementite. Cementite has a carbon content of 6.67 per cent.138 The fron-Carbon Equilibrium Diagram Fig. 106. Microstructure of iron: a—territe, %500, 6—austenite, x 250 Fig. 107. Schematic representations of crystal lattices: @—austentte, 6—cementiteThe Iron-Carbon Equilibrium Diagram 139 Cementite has an orthorhombic lattice with close-packed atoms (Fig. 107, 6). Its lattice constants are: a=4.518 A and c=5.069 A. Six atoms of iron form an octahedron with the carbon atom at its centre. The bond of the lattice is of the metallic type (Fig. 107, 5). The melting point of cementite has not been exactly determined but is approximately equal to 1,550° C. Up to a temperature of 217° C, cementite is ferromagnetic. A characteristic feature of cementite is its high hardness (BAn2>700) and low ductility. Solid solutions may be formed on the basis of cementite but the limits of solubility have not yet been established. Under definite conditions, cementite decomposes to form free carbon as graphite. . 7-2. The Iron-Carbon Equilibrium Diagram A modern version of the iron-carbon equilibrium diagram is illustrated in Fig. 108 *. The iron-carbon equilibrium diagram concerns transformations that occur in alloys having compositions from pure iron to cementite (6.67 per cent C). Point A (1,539° C) on the diagram is the melting point of pure iron and point D (~1,550° C) is the melting point of iron carbide (cementite, Fe,C). Points N (1,401° C) and G (910° C) correspond to the allotropic transformations of a-iron into y-iron. Point E shows the solubility limit of carbon in y-iron at 1,130°C (2.0 per cent). The upper left-hand portion of the diagram (Fig. 108) represents the allotropic transformations yata(6) at high temperatures. At temperatures on the line AB (Fig. 108), crystals of the solid solution of carbon in a-iron (6-solution) begin to precipitate from the liquid alloy. The line HJB represents peritectic transformation taking place at a constant temperature, The result of the peritectic reaction is the formation of the solid solution of carbon in y-iron, i. e., austenite. To simplify further study of the iron-carbon diagram we shal! ignore this part of the diagram in the future since it is of no practical importance. The alloys begin to solidify when they reach a temperature corresponding to the line ACD (tiquidus) ** (Fig. 109). Complete solidification corresponds to the solidus temperatures AECF. Austenite is precipitated from the liquid alloy along line AC and cementite along line CD. At point C with a temperature of 1,130°C and at * Only the metastable equilibrium diagram of iron-carbon alloys (Fe-FesC) will be considered in this section. The stable diagram (Fe-graphite) is shown in Fig, 247. #7 Sce the simplified diagram.GreBeip upg mbes woqres-uom ay) “E01 “By 05 The _iron-Carbon Equilibrium Di 2 CEE TS || abit A | | Nts PLEA ET a J CCCoe_ The /ron-Carbon Equilibrium Diagram iat a carbon content of 4.3 per cent austenite and cementite are simultaneously precipitated from the liquid alloy to form a eutectic called tedeburite. All all containing. more tl 2.0 per cent C (cast irons) and completely solidifying at 1,130°C (line ECF) form: the eutectic (ledeburite). . % carter _ Fig. 109. Simplified fron-carbon equilibrium diagram : A discussion of the primary solidification (transformation from othe liquid to the solid state) of steels containing 0.76 per cent C and ‘cast irons containing 3, 4.3, and 5.75 per cent C will clarify the in- Aetgretation of the diagram. . f teel containing 0.75 per cent C begins to solidify at the temperature ¢, (Fig. 110). Here the solid solution of earbon in y-iron, austenite, begins to precipitate from the liquid alloy. The amount142 The fron-Carbon Equiltbrium Diagram of austenite continuously increases as the temperature falls and " its composition varies along the solidus (AE) while that of the liquid phase varies along the liquidus (AC). For example, at a temperature #,, the composition of austenite corresponds to point a; that of the liquid phase {s indicated by point b. This steel completely solidifies at temperature f, and will consist only of austenite. Steel of any composition will solidify in exactly the same’ manner and will consist of only austenite, at the moment when the process has been completed. te Fig. 110. The iron-carbon equilibrium diagram (primary crystallisation) Cast iron containing 3.0 per cent C begins to solidify at temperature f, lying on the liquidus (AC). Here too austenite crystals precipitate from the liquid alloy in a continuously increasing amount as the temperature falls. The composition of the austenite and of the liquid phase may be determined at any temperature, for example, #, by drawing a line through this temperature parallel to the abscissa so that it intersects the liquidus and the solidus. Point d indicates the austenite composition and point e, the composition of the liquid phase which is in equilibrium at lemperature ¢,. The proportion of the amounts of the liquid and solid phases at any temperature between the liquidus and solidus is determined by the lever rule. Thus, at the temperature ¢, the quantitative proportion of these phases is expressed by the ratio bd te Cast iron containing 3.0 per cent C completely solidifies at temperature f, (1,130° C). At this temperature, the liquid phase, enriched in carbon up to 4.3 per cent (the eutectic concentration), is finally solidified. From it, saturated austenite crystals (2.0 per cent C) precipitate simultaneously with cementite to form the eutectic mixture, ledeburite. Therefore, directly following solidification, the castThe fron-Carbon Equilibrium Diagram 143, iron will consist of primary austenite crystals and ledeburite, formed at 1,130° C (4,). Cast iron of any composition between 2.0 to 4.3 per cent C (hypoeutectic cast irons) will solidify in exactly the same way as has been described above. An alloy containing 4.3 per cent C (eutectic cast iron) differs from the previous alloys in that it solidifies at a constant temperature (point C at 1,130° C). Above this temperature, the alloy will be in a liquid state and when the temperature falls to 1,130°C {point C), the alloy solidifies at this constant temperature forming ledeburite, Soliditication of the eutectic alloy at a constant temperature conforms to the phase rule. In a two-component system containing three phases (liquid phase, austenite, and cementite) F=2+1—3=0, i, e., the system is nonvariant. All three phases have definite compositions (liquid phase--4.3 per cent C, austenite—2.0 per cent C, and Fe,C—6.67 per cent C) and the transformation temperature is constant—1,130°C. A horizontal step on the cooling curves (Fig. 110) - corresponds to the formation of the eutectic (ledeburite). Cast iron containing 4.3 per cent C will consist ofonly ledeburite after solidification, The solidification of Aypereutectic cast irons (4.3 to 6.67 per cent C), as is evident from Fig. 110, begins along line CD by the precipitation of cementite (Fe,C) from the liquid alloy. A cast iron containing 5.75 per cent C, for example, begins to solidify at the temperature ft, where cementite crystals (Fe,C) begin to precipitate from the liquid alloy. Since cementite is the phase richer in carbon, its precipitation will change the composition of the liquid alloy, upon a further fall in temperature, toward a reduction in carbon content along the liquidus CD. At 1,130° C, the liquid alloy will reach the eutectic composition (4.3 per cent C) and will solidify at this (constant) temperature to form ledeburite. The completely solid alloy will consist of primary cementite crystals and ledeburite. All hypereutectic alloys will solidify in a similar manner. Thus, all steels consist of austenite after the primary crystallisation, while all cast irons consist of ledeburite and primary crystals of excess austenite (hypoeutectic cast irons containing from 2.0 to 43 per cent C) or cementite (hypereutectic cast irons containing from 4.3 to 6.67 per cent C}. We will next consider transformations known as secondary crysta- Ilisation which occur in the solid state. In the iron-carbon alloys, these are associated with the transformation of y-iron to a-iron and the decomposition of austenite. The line GS in the equilibrium diagram (Figs. 109 and 111) indicates the beginning of austenite decomposition and the precipitation of ferrite from the austenite. The critical points along the GS line are designated as Ac, in heatin; and as Ar, in cooling. The line SE indicates the temperatures a which austenite begins to decompose with the precipitation of excess carbon as cementite. This latter is called secondary or proeutec-

Engineering Physical Metallurgy by Y. Lakhtin

Uploaded by

Copyright:

Available Formats

Engineering Physical Metallurgy by Y. Lakhtin

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Engineering Physical Metallurgy by Y. Lakhtin

Uploaded by

Copyright:

Available Formats

You might also like