Materials Science and Technology

An X-ray fluoroscopy technique has been used to study bubble formation and blocking in horizontal nozzles during the injection of nitrogen gas into liquid cast iron. Bubble frequency and mode of formation were found to be insensitive to the gas flowrate over the range investigated. Blocking is believed to be associated with the periodic collapse of bubbles at the nozzle exit, causing metal droplets to be propelled into the nozzle tube. Nozzles designed to reduce blocking were examined.

An investigation has been made of the effects of (a) solution-treatment temperature, (b) cooling rate after solution treatment, (c) aging treatment, and (d) composition (i.e. molybdenum and silicon contents) on the microstructure and mechanical properties of some α-alloys of the Ti–Al–Zr–Mo–Si system related to IMI 685 Ti-alloy composition (Ti–6Al-5Zr–0·5Mo–0·25Si wt-%). β-solution treatment following quenching produces martensitic structures which are accompanied by small amounts of retained β-phase in the molybdenum-containing alloys. Increasing the β-solution treatment temperature does not significantly affect the martensite plate size but increases the original β-grain size, without altering markedly the strength. Reducing the cooling rate after the solution treatment increases the amount of retained β-phase and changes the microstructure from a martensitic- to a basket-weave-type morphology. The strengths of these alloys are enhanced by dynamic strain aging (DSA) in the range ∼700–850K, and this is considered to derive from an interaction between silicon and interstitials which produces a strong interaction with mobile dislocations. Reducing the cooling rate after solution treatment and prolonged aging treatments remove silicon/rom the matrix solution and reduce the DSA effect.

An investigation has been made of the effects of alloy composition and of heat treatment on the high-temperature creep resistance of some α-alloys of the Ti–Al–Zr–Mo–Si system related to the IMI 685 Ti-alloy composition (Ti–6Al–5Zr–0·5Mo–0·25Si wt-%). At least two mechanisms are rate controlling in the temperature range 660–860 K. At low temperatures in this range the creep exhibits nearly a thermal characteristics, the rate being only slightly dependent on temperature and characterized by very low apparent activation energy values ∆H_ap and by some anomalous values of stress exponent n. The behaviour is attributed to a drag creep mechanism. The high values of ∆H_ap and n observed for creep at higher temperatures are attributed to microstructural changes which occur during the tests. Stress-assisted precipitation occurs on dislocations during creep and it is proposed that the creep rate is limited by the rate at which the dislocations can break away from the precipitates.

The influence of composition on the tendency for surface-crack propagation to occur on straightening of continuously cast strand of a commercial C–Mn–Nb–Al grade of steel has been examined and compared with the hot-ductility behaviour obtained from Gleeble tensile tests, using reduction of area as the measure for hot ductility. The tendency for surface-crack propagation was found to increase with Nb and soluble Al addition and to decrease with a rise in the P level. Hot ductility was also found to be affected in a similar manner by these elements. The results have been interpreted in terms of differences in the effectiveness of the NbCN precipitation in pinning the γ grain boundaries during deformation. From the close agreement between the hot-ductility behaviour as given by the Gleeble tests and the tendency for surface cracks to propagate during straightening of these steels at the exit from the continuous-casting machine, it has been concluded that the Gleeble test can be used with a fair degree of confidence to predict the behaviour on straightening of other steel compositions made by the continuous-casting process.

The very gradual rise in the residuals content of scrap is reflected in the levels of residuals found in engineering steels. A superficial survey of national standards and a few ‘in-house’ specifications shows considerable variation in the limits imposed on residual contents. The normal levels in UK steels are so high that for some specifications purer and more expensive raw materials must be used; this is reflected in the increased price of ‘low-residual’ steels. One way of counteracting this trend is to restrict the demand for low-residual steels to applications where they are essential. The most serious detrimental effects of high residuals contents are hot cracking and the incidence of surface defects. The critical contents of Cu and Sn depend on the process route and final application. Steel users should specify residual levels in line with the ‘enrichment factor’ applicable to their processing route. Residuals may raise the flow stress at room temperature and thus increase tool loads and tool wear in cold working. Tool-load measurements have shown a close correlation with hardness. Estimates of the effect of residual elements on hardness increment indicate that current melt-out levels correspond to a hardness increase of about 20 HV or an increase of 80Nmm⁻² in UTS. This justifies the demand for low residuals in material subjected to severe forming operations. However, steel users should give serious thought to the actual forming operation before specifying low residuals. The effect of residuals on hardenability is beneficial. The engineering industry in the UK derives considerable benefit from the ability to obtain good mechanical properties by oil quenching what are officially classed as carbon steels. A number of steps which can be taken over the coming years to cope with the rise in residuals are noted.

The interphase boundaries in grey and ductile cast iron were studied with a scanning Auger microprobe. Sulphur and oxygen were found to be adsorbed at the flake/metal interfaces in the grey iron, while the nodule/metal and intercrystalline graphite interfaces in the ductile iron were free of foreign elements. The only magnesium detected in the magnesium-modified ductile iron was combined with phosphorus and sulphur as a compound. A model is presented in which it is proposed that Fe–C–Si eutectic alloys in the absence of surface-active impurities (e.g. in vacuum casting o/high-purity materials) produce nodular graphite owing to the inherent instability of the graphite/melt interface. The sulphur and oxygen always present in commercial alloys adsorb at the graphite/melt interface, effectively ‘stabilizing’the active sites on the graphite basal planes, and preventing spherulitic growth. The purpose of modifiers is to getter these impurities.

The continuous casting of both ferrous and non-ferrous metals has been investigated extensively over the last 20 years, reflecting the major position this casting technique has now assumed in industry. Recent reviews have dealt with the ∼ontinuous casting of steel and aluminium. No attempt is made to cover the subject in a comprehensive way as it is too large. Instead, particular areas are discussed primarily those relating to solidification, mathematical modelling, and cracking.

Some recent important advances in solidification theory and practice are summarized. As solidification technology is advancing so rapidly only a representative sample of topics are discussed, including ingot and shape casting processes, grain size and grain-size control, dendrite structure, and inclusions and microporosity.

In multipass submerged-arc welds deposited from 2· 25 Cr–Mo wires of controlled difference in purity, the reheat-cracking susceptibility was only slightly dependent on the residual-element content. When the wires were used to deposit tungsten inert gas (TIG) welds, the effect of residual elements remained small, but the extent of grain refinement was significantly increased and the reheat-cracking resistance consequently improved. A TIG weld deliberately made to maximize grain refinement was of very high resistance to reheat cracking even though of relatively low purity. Hence, it is concluded that by appropriate choice of welding process, grain refinement can be maximized in both weld metal and heat-afrected zone, and reheat cracking in low-alloy power-plant steels consequently can be prevented. Special measures to reduce the residual-element content below the levels generally obtained in current low-alloy steelmaking practice, therefore, should be unnecessary.

An electronic switch circuit is described for a capacitor spark-discharge hardening process which permits hard coatings to be deposited on suitable substrates at coverage rates of up to 9cm² min⁻¹, about an order of magnitude higher than by conventional mechanical switching means. The process was evaluated in a hot-forging die-life programme, where it was found that the life of extrusion pinion dies could be increased significantly (> 10%) by the coating process.

In the rolling of sheet and strip in modern mills, deformation takes place mainly in the longitudinal direction. As the extension increases, there is a deterioration in the transverse and through-thickness toughness parameters, attributable mainly to the plastically deformable inclusions, such as manganese sulphides. The form of the inclusion in the wrought product is a function of its morphology in the cast steel and its plasticity. The effect of composition on both factors has been investigated. Increasing the aluminium or sulphur content of the steel promotes the formation of Type II manganese sulphides, which are precipitated as extensive arrays of rods in an interdendritic eutectic distribution, and which deform, as an array, to produce highly elongaled stringers. Mildly deoxidized steels contain globular, isolated, Type I sulphides, which do not form the particularly harmful stringers, but which are still highly deformable. Cerium and calcium additions successfully modified the initial morphology of the sulphide inclusions to an isolated globular form, and also decreased their plasticity, whereas titanium additions of up to 0·12% were ineffective in these steels at the manganese levels involved. The electron microstructure of deformed α-MnS, α-Al₂O₃, and Y-phase inclusions have been investigated. Evidence of recovered substructures, recrystallization and precipitation within manganese sulphides, and Ti₃S₄ precipitation within V-phase inclusions is presented.

The effects of trace amounts of Pb, Sn, Se, Te, and Tl on the creep and stressrupture properties of Nimonic 105 at 815°C have been investigated. The work confirmed that with the exception of Sn, all the elements have a deleterious effect when present in a sufficient amount. The most harmful impurities were Te and Pb. A generally more sensitive indication of the presence of harmful amounts of impurity was obtained with notched specimens. Optical metallography established that the deleterious effects were associated with greatly increased intergranular cavitation which became evident at an early stage of creep life when impurities were present. Auger electron spectroscopy showed that about one monolayer of lead was present on a cavitated grain boundary in a creep-tested specimen containing 34 ppm of Pb.

The effects of boron, titanium, and nitrogen contents on the hardenability of boron-bearing steels have been studied. The boron hardenability factor was found to be a rapidly decreasing function of (total N – N as TiN). Thus, it was confirmed that titanium plays a very important role in retaining the excellent hardenability through its scavenging effect for dissolved nitrogen. From the results of this work, a useful guide concerning the limitation of total nitrogen content was derived for the development of titanium-free boron steels characterized by a good combination of hardenability, machinability, and toughness, especially for heat-treated parts made for heavy equipment.

The ballistic penetration behaviour of a high-strength aluminium alloy, 7001 T6, was investigated and targets of the alloy were found to fail in two modes, discing or plugging, depending on the orientation of the target in relation to the extrusion direction of the round bar from which it was cut. The dependence of failure mode on orientation was related to poor fracture toughness in planes containing intermetallic stringers which were aligned parallel to the extrusion direction. A model for discing failure was developed which allows the penetration resistance of the plates to be calculated from the target strength properties. It is possible to prevent both the discing and the plugging failure modes by the use of steel back-up plates in contact with the rear of the target.

A study has been made of the adherence of plasma-sprayed aluminium coatings to a wide range of metallic substrates to examine whether the mechanism previously put forward by the authors for steel substrates is of general application. It was established that aluminium adheres to metallic substrates by one of two mechanisms. At high substrate preheat temperatures, adhesion is generally by the formation of a metallurgical bond. At the lower preheat temperatures normally used, the presence of an adherent substrate oxide film with which the aluminium can react is necessary for adequate adhesion.

A new process for producing rapidly chilled metallurgical microstructures under reproducible cooling-rate conditions has been developed. Processing is accomplished by rapidly traversing a continuous, high energy-density laser beam over the material to produce a very thin molten layer at the material surface. The high specific rate of energy delivery facilitates surface-localized melting at very high melting efficiencies, i.e. the major portion of the absorbed energy is used for melting, with only a very small fraction going into heating of the solid, subsurface material. This ability to maintain a cold substrate while melting a thin surface layer of material results in rapid quenching of the molten layer once the energy source is removed. Calculations made for pure nickel indicate that cooling rates of 5 × 10⁸ Ks⁻¹ are attainable in layers that are 2·5 × 10⁻³ mm thick, while 2·5 × 10⁻² mm thick layers cool at rates of up to 5 × 10⁶Ks⁻¹. The importance of rapid quenching is based on the fact that the unsurpassed homogeneity of the liquid can be preserved (or nearly preserved) in the solid, which can then be utilized in that form, or these structures can be further modified in order to produce other desired properties. The subject process, termed’ laser glazing’, has been shown to be capable of producing a wide variety of novel and interesting metallurgical microstructures on the surfaces of bulk metals. Amorphous surface layers have been produced on bulk substrates by laser glazing compositions at or near deep eutectic troughs. In other compositions, extended solid-solution phases have been obtained. These can be decomposed into finely divided multiphase microstructures in the solid state. Such compositions are candidates for new dispersion-strengthened alloys. At higher second-phase contents, extremely fine filamentary eutectic microstructures have been obtained. A strong tendency for oriented overgrowth (epitaxial growth), in various orientations, has been demonstrated for a variety of mono crystalline substrate materials. Laser glazing (in conjunction with surface alloying, post-glaze heat treatment, and/or deformation) has great flexibility as a means to generate a broad range of novel surface microstructures and properties. Furthermore, although currently being employed primarily for the surface treatment of materials, it also has the potential for processing bulk materials by the sequential build-up of one glazed layer upon another.

A die-casting system is described which utilizes the thixotropic nature of semisolid Rheocast slurries. This system has been developed to a pilot-plant scale at the Massachusetts Institute of Technology. This plant comprises a continuous Rheocaster which is used to produce ingot stock, a reheating furnace to reheat individual charges to the semisolid state, a ‘softness indicator’ to determine when the reheated charges are suitable for casting, and a commercial cold-chamber diecasting machine. This paper deals primarily with the structure and properties of Thixocast ferrous alloys. Discussion is given of control of primary particle size, phases, oxide inclusions, and microsegregation in AISI 304, 440C, and other steels. Structure–property relationships are also presented and comparisons made with the properties of conventionally cast and wrought alloys.

A series of ingots was produced containing varying quantities of hydrogen and non-metallic (oxide) inclusions from melts of super purity aluminium. The incidence of blistering in sheet and foil was evaluated in terms of hydrogen content, melt cleanliness, thermomechanical treatment, and lubrication during rolling. Tensile properties were used to monitor the quality of the material. The importance of low hydrogen and inclusion contents and of maintaining a good surface quality during processing and the necessity for lubricant removal before annealing have been established as crucialfactors in the avoidance of blistering.

The microstructure of the built-up edges of a number of alloys has been examined using both optical and electron microscopy. The built-up edges of the two ferrific steels examined showed that these were significantly different to those formed in austenitic stainless steel and Cu–40Zn brass. The built-up edge deformation system has been described using three zones: primary shear (1), deformation around the build-up (2), and deformation to form the build-up (3). The ferrific steels exhibited high-strain deformation structures in both zones 2 and 3 with structures similar to those seen in wire drawing and swaging. The stainless steel and Cu–40Zn brass, however, showed equiaxed grains about 0·2μm in size with little or no twinning in zone 3. Zone 2 of these materials showed considerable twinning and the resulting surfaces were markedly less torn than those oftheferritic steels.

An examination has been made of alumina formed during the deoxidation of iron–oxygen alloys by aluminium, and also of the fracture of the alumina during mechanical working. The morphology of alumina is controlled largely by the supersaturation of the melt, high supersaturation giving dendritic alumina while low supersaturation gives liquid mixed oxides which react to form alumina and spinel reaction rims. The morphology of the alumina formed direct from the melt is controlled largely by the dissolved oxygen content which, in turn, controls the alumina/melt surface energy. Spheroidization and sintering of the dendritic alumina occurs as the dissolved oxygen of the melt decreases until dense, sintered alumina aggregates are formed. During deformation, the alumina becomes fractured and disseminated into stringers. Dendritic alumina fractures by breaking off the dendrite arms, while the polycrystalline alumina aggregates fracture intergranularly. Mechanisms are suggested for the formation of the various forms of alumina, and the processes involved in their fracture are discussed.

Studies have been performed on the effect of five minor elements (S, Ca, Mg, Y, and Zr) on the hot workability of three solid-solution strengthened, nickel-base superalloys: Inconel 600, Hastelloy-X, and a Ni–Cr–W type alloy (NSC-1) for heat-exchanger pipes in high-temperature gas reactors. The hot workability was evaluated from the reduction of area in hot tensile tests using a Gleeble testing machine. Round bar specimens were prepared from near surface portions of alloy ingots melted and cast by vacuum induction melting, vacuum arc remelting, or electroslag remelting. When the sulphur content was increased, a ductility dip appeared for each alloy between 950° and 1150°C; the greater the sulphur content, the deeper the dip. At the same time, the fracture appearance became increasingly intergranular. By adding a small amount of Ca, Mg, Y, or Zr, the hot workability of these alloys changed drastically, and it was found that controlled addition of these minor elements gave a superior hot ductility. Defining ∆S as ∆S=%S–0·8 × %Ca–0·3 × %Mg–0·5 × %Y–0·1 × %Zr (wt-%) the hot ductility of these alloys changed systematically as follows, irrespective of the three refining processes: the ductility dip diminished and excellent ductility was achieved with 0·003>∆S>–0·004. The hot ductility decreased gradually with ∆S<–0·004, and became extremely poor when ∆S was larger than 0·003. Investigation of fractured surfaces with Auger electron spectroscopy and electron probe microanalysis showed that when ∆S was positive, atomically segregated sulphur was present on the intergranular fractured surfaces, while when ∆S was negative, atomic sulphur was not detected, but metallic sulphide compounds were observed both in grain interiors and on grain boundaries. The hot workability of nickel-base superalloys, therefore, can be related to the ∆S value, and in order to improve the hot workability of these alloys, minor elements should be added so that ∆S becomes nearly zero. By the controlled addition of Ca, Mg, Y, and Zr, a newly developed Ni–Cr–W type alloy (NSC-1) has been successfully processed.

The surface composition of swarfs from leaded Cu-40Zn has been determined by Auger electron spectroscopy and ion etching as a function of bulk-lead content and machining conditions. The lead additive is distributed at the swart surface as a continuous thinfilm of monatomic thickness and as varying proportions of thick (p.m) lead striations along the cutting direction. The surface lead concentration of the chip is much higher than in the bulk metal. There is also a strong enrichment of zinc at the swart surface. It is considered that the lead surface layers are formed by a process of extrusion of lead owing to the high compressive forces which are developed during cutting. One complete lead monatomic layer is formed at bulk contents≥ 0·88 wt-% and this is associated with the absence of a secondary-shear zone, pointing to the lubrication of the tool/swarfinterface by the lead. A lead content in excess of 2·76 wt-% reduces power consumption, produces discontinuous chips, and is sufficient to maintain a lubricant monatomic layer of lead and 30–50% coverage of thick lead at the swarfsurface over a very wide cutting speed range of 2·5–200mmin⁻¹. The swarf-surface lead concentration is unaffected by the presence of external oil–water lubricants except at the lowest speed.

After outlining the origins and early developments of copper metallurgy in Poland, the growth of the industry since the end of the Second World War is described, with emphasis being laid on the development and adaptation of machinery and techniques to the particular Polish requirements. Examples of research and development are given from the areas of mining, ore upgrading, copper-concentrate smelting, and copper and copper-alloy processing. To illustrate the types of research being carried out in different subject areas, the author describes some of the aspects of studies being made of the liquid state and crystallization, crystallization in the continuous-casting process, working and recrystallization, and phase transformation.

The interaction of dislocations with precipitates in an age-hardened polycrystalline alloy (the total amount of alloying elements is about 2 wt-%) is reported. An electron microscope study of thin foils aged similarly to the tensile-test specimens reveals that during aging between 450° and 600°C a precipitation sequence exists: GP zones – metastable coherent particles – non-coherent particles of a stable phase. The presence of GP zones and coherent particles increases the yield stress but has little effect on work-hardening characteristics. Precipitation of non-coherent particles initiates high work-hardening rates. The high yield stress of the zone-containing alloy is owing to the work done in forcing the dislocations through the zones. The high work-hardening rate of the specimens containing non-coherent particles suggests that the dislocations do not cut through particles, but avoid them.

Inhibition of recrystallization in copper and its alloys can be caused by small quantities of foreign elements such as Ag, P, Cd, Mg, and Sn in solid solution in the copper. These elements segregate at dislocation and grain boundaries, thereby inhibiting migration. The effect of cold work is a simple function of the quantity of soluble addition and the degree of cold work. The situation is more complicated for precipitation-hardened alloys because cold work, precipitation, and softening interact. In starting with the supersaturated solid solution which has been severely cold worked, heat treatment will lead to both precipitation and recrystallization. However, by means of precipitation treatment before cold working, it is possible to slow down, and even halt, normal recrystallization; this treatment leads to the formation of a critical distribution of precipitates (‘dispersoid') which effectively halts both dislocation and grain-boundary movement. Cu–Ti, Cu–Zr, Cu–Fe, and Cu–Co alloys were studied for different sequences of heat treatment and cold working and the following characteristics are described: (a) a very small residual quantity of soluble Zr suffices to maintain the softening temperature above 400°–500°C; (b) soluble Cr also inhibits the softening of Cu, although larger amounts are required than for Zr,’ (c) fine precipitates of Cr in the form of a critical dispersion are also effective for retaining cold working; (d) iron in solid solution has practically no effect on the softening temperature of copper.

The tensile properties, impact properties, and microstructures of a 0·09C–1·25Mn–0·02Nb steel have been measured in 13 mm plates cooled at rates of 1·8–26 Ks^−l after controlled rolling at 650°, 800°, and 1000°C with rolling reductions of 50, 67, and 83%. At finish-rolling temperatures of 800° and 1000°C, cooling rates of 1·8–6 Ks^–l raised the yield stress 20–60 N mm⁻² above the value for air-cooled plates, without impairing the impact properties. Further increases in yield stress at a cooling rate of 26 Ks^−l were accompanied by a decrease in impact toughness probably as a result of the presence of about 10% martensite in the microstructure, and a susceptibility to quench aging and strain aging developed. Lowering the finish-rolling temperature to 650°C produced yield stresses of about 550 N mm⁻² for cooling rates of 1·8–26 Ks^−l compared with 441Nmm⁻² for the air-cooled condition, and the impact properties were not impaired provided the cooling rate did not exceed 6 Ks⁻¹ and the rolling reduction was 67% or higher. The improvement in properties and the low dependence of properties on changes in processing parameters at rolling temperatures of 800° and 1000°C could make accelerated cooling commercially attractive for steels with suitable compositions.

An alternative form of the hyperbolic sine relationship is developed for use as a constitutive equation at elevated temperatures. It is shown how an optimization technique may be applied to results from torsion tests to determine the constants in this equation and that this method must be more accurate than those presented previously. Experimental results for two aluminium alloys indicate that the activation energy during hot torsion testing is the same as the activation energy for bulk self-diffusion. It is shown that thermal changes occurring during testing affect results significantly.

The importance of determining the specimen temperature at varying strains and strain rates during the twisting of a torsion specimen is emphasized and it is argued that previous analyses are inadequate. A three-dimensional finite-diflerence model is developed and a method of obtaining an ‘average’ temperature rise is presented. It is shown that significant temperature profiles are set up during torsion testing and that axial conduction cannot be assumed to be zero. The assumption of a laminar boundary layer is acceptable. Results are presented for two aluminium alloys.

The Cu–Mg–P system has interesting properties. A detailed study of the precipitation process and its effect on the physical properties has been made. The solubilities of magnesium and phosphorus in copper are rather high. Solubility is reduced drastically in the ternary system through the formation of Mg₃P₂. The shape of a quasi-binary section Cu–Mg₃P₂ is proposed. Segregation during solidification is pronounced and results in a complex mixture of non-equilibrium phases. The kinetics of the precipitation process have been studied. The rate of precipitation is slow in a recrystallized structure but faster after cold working. The precipitates dissolve rapidly when the alloy is heat treated to temperatures above 600°C. Finely dispersed precipitates give a hardening effect and restrain recrystallization. It is possible to obtain very high tensile strength and a conductivity of 70%10 IACS by a combination of precipitation and cold deformation.

Rates of accumulation a and annihilation β of dislocations, defined in the differential equation dρ/dδ=α/bd–βρ have been evaluated by determination of the true stress–true strain curve of the metal, direct measurements of dislocation-cell diameter d, and determination of dislocation density ρ by a calorimetric method.

The stretchability of strip-cast and DC-cast Al–Mn alloys as measured by their forming limit curves has been studied. Optimization seems to be achieved with as high a work-hardening exponent as possible in uniaxial tension when instability is approached. This work hardening rate is, in the present case, influenced more by the slope of the stress-strain curve than by the flow-stress level which also has an effect. At these high strains the increase inflow stress is, in general, proportional to the square root of the dislocation density. Consequentially, both the rate of production and the annihilation and polygonization of dislocations have an important effect on the work-hardening rate. The strip-cast and non-homogenized alloys have the lowest work-hardening rates of the alloys investigated. They contain a large number of finely dispersed secondary particles which give rise to an irregular and tangled dislocation structure after straining to maximum load in uniaxial tension. It is suggested that this kind of structure facilitates a more pronounced dislocation annihilation than one consisting predominantly of cell walls. In general, the observed da/dE v. strain curves at large strains increase with decreasing dispersion ratio (volume fraction/average radius) of the particles.

A large number of ferrite–pearlite steels have been examined and measurements made of grain size, pearlite volume fraction, degree of precipitation hardening, and thickness of grain-boundary carbides, the latter being measured using a scanning electron microscope. Multilinear-regression analyses were carried out on these measurements and on the composition of the steels, to obtain statistically the best structure–property relationships for 500;0 FATT, and 54J and 27J Charpy impact transition temperatures. Regression equations based solely on the structural parameters, grain size, precipitation hardening, and carbide thickness were found to give improved predictions over previous regression equations in which no factor for carbide thickness has been included. The importance of carbide thickness in influencing the impact properties was substantiated using data from pearlite-free steels in which the major variable was the carbide thickness. The carbide thickness was found to be dependent primarily on the manganese content and cooling rate from the austenite temperature range, an increase in either leading to finer carbides and improved impact properties. At the lower manganese levels (<1%Mn), micro-alloying additions also were found to influence the carbide thickness.

A comprehensive search has been made for alloying elements which could be effective decolourizers in an 18 carat (i.e. 75 wt-% gold) alloy. Although 16 potentially suitable elements were identified, assessment of experimental binary alloys revealed that only two of these elements, Ni and Fe, were acceptable as major alloying elements in an 18 carat, castable white gold jewellery alloy having a combination of properties which manufacturing jewellers currently believe to be ideal. The major shortcoming of Au–Ni alloys was that they were hard and work hardened rapidly but these disadvantages could be overcome effectively by adding cobalt. More complex 18 carat golds based on Au–Ni–Co alloys were designed using the information previously derivedfrom the binary gold-based alloys. Subsequently, alloys in the composition range 75Au–7–1 2Ni–7–12Fe (with Ni+Fe = 16−19)−2−3Co−2−3Cr−2−3In (in wt-%) were identified as having an improved combination of colour and hardness over those of existing commercial, castable 18 carat white golds.

By applying the combined effusion vaporization from a Knudsen cell and mass spectrometry sensing techniques the thermodynamic properties of liquid Cu–Ag solutions were investigated. The method of measuring the intensity of the ⁶³CU₂₉ and ¹⁰⁷Ag₄₇ ion currents J is described. The enthalpies of vaporization and of sublimation of these metals were estimated from the temperature dependence of the ion-current intensity values. The investigation of the Cu–Ag system was carried out over the range 1350–1550 K for ten selected compositions covering the complete concentration range of the system explored. The obtained dependence of J+⁶³ CU ₂₉/J+¹⁰⁷ Ag ₄₇ on temperature served to determine the activity and the coefficients of Cu and Ag activity by means of the modified Gibbs–Duhem equation. The partial thermodynamic functions of Cu and Ag in the solutions examined were also determined. These experimental values are compared With the data obtained by other authors.

Using an emf technique in concentration cells with solid electrolytes in the scheme Pt, Ni+NiO/ZrO₂+CaO/O in liquid Cu/Cr cermet,Pt, oxygen activity in liquid copper at temperatures of 1100° and 1200°C was determined. The oxygen content of liquid copper was varied by addition of solid CU₂O. The results obtained show that at concentrations of oxygen up to 0·5 at.-%, Henry's law is valid. The results on the limiting activity coefficients of oxygen at 1100° and 1200°C and on the standard free energy of oxygen dissolution in liquid copper show good agreement with published data.

The structure of extruded wires of three grades of copper with purity higher than 99·7% has been studied. The structure is controlled directly by extrusion temperature and cooling velocity after the dynamic recrystallization of materials at the die outlet. It is possible to prevent both uniform and selective grain growth by secondary recrystallization, and grain coarsening by decreasing the extrusion temperature and intensive cooling of the extruded wire. The structure is not determined by grain orientations in the cast billets. It was found impossible to remove the non-uniformity of the extruded wire structure by recrystallization, even after drawing with a 40−50% reduction. A uniform recrystallized structure of highpurity copper is obtained only in wires drawn with a reduction greater than 80%. The lower extrusion temperature in the range 830°−880°C results in a more uniform structure after a lower degree of cold work, and thus in higher toughness expressed by elongation.

During creep exposure of austenitic stainless steel, structural changes take place; these are considered to be microstructural aspects of creep damage and have a negative effect on the steel's mechanical properties. Structural changes were studied extensively for one particular heat of steel AISI 304 and, afterwards, related to its deterioration in mechanical properties.

An existing steel was modified to a higher strength level to meet the requirements of airframe structures. Factorial-regression analysis of data determined previously helped to determine the interaction effects of the selected alloying elements on mechanical properties. The range of alloying elements required to optimize toughness at the 1586–1724MNm⁻² ultimate strength level was subsequently determined by experiment. The developed alloy 0·16C–2Cr–1Mo–10Ni–14Co (AF 1410), attained a fracture toughness K _IC = 156MNm^−3/2 at this strength level. These properties result from the precipitation of M₂C-type alloy carbides in a highly dislocated, lath martensite matrix during aging.

The role of lead in free-machining brass has been studied by calculating the temperatures formed in the primary and secondary shear zones of chip formation and by examining, in a scanning electron microscope, the structure of chips from cutting experiments. The calculations show that the temperatures in the primary zone do not exceed 200°C for workpiece speeds up to 200mmin^−l. The discontinuous chip segments are formed as ductile fracture occurs along a shear instability in the primary zone. Scanning electron microscopical observations of the cuplets on the free surface of the chips suggest that voids occur at the lead inclusions so that the void–sheet mechanism of crack propagation gives rise to the discontinuous chip. A possible cause of void initiation at the lead inclusions is discussed. In the chip/tool contact region the major function of the lead is to provide layers of low shear strength within the secondary shear zone. Lead is also drawn out of the many voids to be deposited on the tool faces. At high cutting speeds the maximum temperatures, occurring at the end of the contact length, exceed the melting point of lead. However, in the speed range investigated this has only a marginal injluence on the chip/tool interaction. Thus, the ‘internal lubrication’ effects, arising over the first part of the contact length, remain the domi/Ulnt function of the lead inclusions in the secondary shear zone.

Some metallurgical processes used to obtain suitable materials for deep drawing (textures with a predominance of {011} components parallel to the sheet surface) lead in the case of thin sheets, to the formation of non-advantageous texture and to excessive material hardness as well as to undesirable grain growth. In order to find suitable production methods for radiator material the quality of sheets obtained by middle temperature annealing and a wide range of final reductions was investigated. Mechanical properties, anisotropy, and hardening coefficients, as well as the sheet structure have been analysed: the grain size and texture were determined qualitatively from pole figures. Two technological processes for obtaining half-hard sheet were used – annealing and rolling. It was found that optimum mechanical properties and plastic anisotropy are connected with the {934}〈223〉 5° texture and with homogeneity of suitable grain size. Rolling with a reduction greater than 85% assures a stable and fully recrystallized texture of this type. It was found too, that more advantageous sheet properties are obtained by annealing to the half-hard state. It was established that the annealing temperature range over 420°C leads to heterogeneity of grain size.

The effects of deformation on the structure of copper and its low stacking-fault energy alloys are reviewed and it is shown that these may be classified according to the method of examination used. In rolled low stacking-fault energy alloys the deformation sequence involves the formation of stacking faults, mechanical twins, and shear bands that are at first restricted to individual grains but subsequently extend from one surface to the opposite surface. Shear bands form at ∼35° to the trace of the rolling plane in longitudinal sections and are composed of very small, slightly elongated crystallites. The twin thickness and crystallite spacing are determined by the stacking-fault energy and decrease as that parameter decreases. In copper the deformation sequence involves the formation of equiaxed cells of dislocations, microbands, clustering of microbands, and shear-band formation. It is shown that measurements of cell thickness as a function of strain in copper are meaningless. Replication and other techniques are used to establish the relationships between the various features of the optical microstructure and the structure seen in transmission electron microscopy. Although there is considerable uncertainty about the nature of shear bands in copper, it is pointed out that shear bands are a common deformation feature in metals and that both shear bands and microbands are manifestations of instability.

Continuous and interrupted torsion tests have been conducted in the range of 900°−1000°C at surface strain rates of 0·1–10s⁻¹ on a low-carbon and a Nb-bearing steel. During continuous deformation, the strains for recrystallization and for fracture had dependencies on temperature and strain rate that were similar for both steels but were higher and lower, respectively, for the Nb-bearing steel. For both steels, the flow stress depended on temperature and on strain rate according to an Arrhenius and a power-law relationship, respectively, and was consistently higher for the Nb-bearing steel. The interrupted deformations were performed so that in a given test, interruptions were of equal duration being 10, 20, or 40 s at equal strain intervals of either 0·2 or 0·4. The difference in stress at the instant of un loading and upon reloading was measured to assess the amount of static recovery and recrystallization taking place during the interval. At 0·1 S⁻¹ recrystallization took place in the carbon steel while recovery operated in the Nb-bearing steel; at 1·0s⁻¹ recrystallization occurred in both steels. The progress of softening was augmented by increase in the strain rate, the temperature, and the interruption time.

Some nickel–chromium alloys have been slit by pressing them against toothless mild steel discs spinning with rim speeds between 20 and lOOms⁻¹. The discs were from 0·13 to 0·76 mm thick and were water cooled. Material was removed from the alloys by adhesive wear. Two states of slitting were observed corresponding to mild and severe wear. The severe state gave slitting times comparable with chip-forming slitting processes, together with acceptably low disc-wear rates. Mild or severe slitting occurred depending on alloy composition, coolant composition, disc speed and load, and test geometry. Alloys with low Ti and Al contents tended to mild behaviour, those with more than 3%Al tended to severe behaviour, and those with a low Al content but more than 1%Ti could be switched from mild to severe behaviour by adding basic sodium salts to the water coolant. Chemical analyses suggest that loss of chromium from the slit surfaces results in mild slitting and suppression of the loss causes severe slitting.

An investigation has been conducted into the contribution made by nitrogen and nitrogen plus carbon to the properties of non-alloy steel treated at 570°C. It is shown that both the fatigue properties of the diffusion zone and the tribological and corrosion properties of the surface compound layer are independent of the carbon content. Recommendations are made on how such processes may be conducted and controlled in practice, and examples are shown of how the property improvements associated with these treatments have been exploited in the design and manufacture of non-alloy steel automobile components.

The solubility of copper in Outokumpu commercial slag at 1250°, 1300°, and 1350°C was studied by achieving equilibrium between the slag-reduced Cu–Pb–Fe alloy and CO–CO₂ mixtures. The results are interpreted by making use of the equilibria

Cu₂O _Slag =2CuO_0·5Slag

Cu₂O _Slag =2Cu _alloy +0·5O_2gas

Polycrystalline copper was irradiated with neutrons up to integrated doses of 6·3×10²⁰ncm⁻² (>0·1Me V), then rolled to a reduction in cross-section of 97%. Single crystals of selected orientations, doped with foreign elements, were rolled up to cross-section reductions of >99%. Some of the undoped rolled crystals were irradiated with neutrons up to integrated doses of 5×10¹⁹ ncm⁻² (>0·1Me V). The heat released during heating of the rolled samples was measured. For the case of the irradiated samples, the recrystallized grain structure, the recrystallization texture, and the migration velocity of individual grain boundaries were determined after various heat treatments. Irradiation with neutrons has the effect of accelerating the overall recrystallization process. The velocity of a given grain boundary increases with increasing neutron irradiation for unimpeded grains. The microstructure, on completion of primary recrystallization, becomes increasingly finely grained with increasing irradiation, but the texture remains unchanged. Also, the recrystallization heat remains unchanged. The effect of the foreign elements on the heat of recrystallization, and on the recrystallization temperature, can be divided into two groups according to their distribution coefficient k: (a) in which the heat of recrystallization is decreased by doping and the recrystallization temperature is increased, and (b) in which the reverse is true. Elements with k<1 belong to the first group, whereas all elements of the second group have k>1. For the case of the pure copper crystals the heat of recrystallization and the recrystallization temperature change considerably with the original orientation of the crystals. The observed results are interpreted in terms of current theoretical knowledge.

High-frequency induction welding of cold-rolled strips to tubes combined with spinner-block drawing is a way of producing very long Cu–10Ni tubes. The results of such a production method are described. The tubes have good surface properties and dimensional tolerances and their mechanical properties and corrosion resistance are similar to those of standard tubes. Some examples of applications are given.

The Uddeholm ferrite-rolling process is a high-deformation process and a thermomechanical treatment. The process has been designed for high-carbon and alloy types of steel strip. The reduction of about 80% is applied at a temperature of 600°−800°C, i.e. above the temperature at which ferrite recrystallizes and below that at which transformation to austenite occurs. The basic innovations concern mechanisms of spheroidization but ferrite rolling is now being used as an effective means of reducing processing costs and throughput times. A joint welding plant produces coils of the order of 10−15 t. A high-frequency heater and a medium frequency heater give short heating times with limited oxidation. Following the inductors are two 4-high stands in tandem, cooling equipment, a flying shear, and two coilers. Hard rolls were needed to resist wear and heat checking. The best solution was to produce a powder-coated compound roll in M3:2.

Three vanadium steels, Fe–0·12C–0·46V, Fe–0·13C–1·5Mn–0·45V, and Fe–0·10C–1·53Mn–0·13 V were cooled over the range 100–0·003 Ks⁻¹. Hardness tests were used to follow changes in strength with cooling rate. A peak hardening was obtained at critical cooling rates in the range 1·5−3Ks^−l for the vanadium steels. The highest peaks coincided with steels of a stoichiometric composition of MC. An increase in the manganese levels in these steels produced a spread in the peak with its consequent decrease and also caused the peak to appear at slightly lower cooling rates. This observation was ascribed in the main to the effect of a lower γ→α transformation on the diffusivity of vanadium and/or carbon.

A series of multipass submerged-arc welds in C–Mn steels produced at interpass temperatures of 60° and 250°C have been studied in the as-welded and stress-relieved conditions. For those welds containing 0·08%C and less than 1·0%Mn stress relieving at 625°C causes an increase in the critical crack-opening displacement (COD) transition temperature, owing to the formation of brittle grain-boundary carbide networks. To explain this carbide development a carbon-concentration profile model, which depends on the extent of the transformation of austenite to pro-eutectoid ferrite is proposed. The results are compared with earlier work on steels containing 0·08%C and more than 1·0%Mn in which stress relieving produced a decrease in the critical COD transition temperature.

The main aim in the paper is to give a brief account of work which has been done in the Institute of Metals Working and Physical Metallurgy, Academy of Mining and Metallurgy, on mechanisms of plastic deformation, dislocation structures, and properties of Cu and α-brasses. The investigations were carried out to determine how stacking-fault energy (SFE), temperature, and strain influence the thermally activated processes olplastic deformation. The work included different tests which were performed over a wide range of temperature and strain rates; these indicated the conditions at which deformation has an athermal character or becomes thermally activated. It was shown that dynamic recovery processes are responsible for differences in flow stress at different temperatures and strain rates and that SFE controls the intensity of these processes.

In order to understand the formation of rolling textures in Cu–30Zn it is necessary to consider the deformation modes which operate at different levels of cold reduction. This has involved a combination of optical microscopy, high-voltage electron microscopy, and X-ray diffraction techniques. There are three essential mechanisms, namely slip, twinning, and shear banding, which are mutually interactive. An evolutionary theory of rolling-texture development which refines and extends existing concepts by incorporating detailed information about the microstructure of the cold-worked state is described.

Crack-growth tests have been made at 565°C on thick-section butt welds produced to assess the effect of manual metal arc process parameters on heat-affected-zone (HAZ) refinement in O·5Cr–0·5Mo–0·25 V steel. It was confirmed that the increased degree of HAZ refinement achieved by weld-procedure control resulted in improved resistance to crack propagation. It was also possible to relate instantaneous rates of crack growth to the local microstructures in the HAZ at the tip of the crack by potential-drop measurements andpost-test metallography. The behaviour of a weld subject to post-weld renormalizing and tempering, as an alternative means of achieving microstructural refinement, was also examined. In this case it wasfound that crack initiation was delayed and the overall growth rate reduced relative to the welded and tempered structures. However, crack growth was accompanied by low displacements, similar to those for an as-welded untempered specimen. These were attributed to the concentration of strain ahead of the crack tip within a narrow, fine-grained decarburized zone adjacent to the fusion boundary.

The abrasive wear resistance and fracture toughness of the principal types of alloy white cast iron have been determined in laboratory tests. It was found that low-stress abrasion resistance is controlled principally by the hard, dispersed carbide phase, and that the hardness of the matrix is of secondary importance only. The strength of the matrix, however, is important in providing mechanical support to the carbide particles. The matrix plays a critical role in determining the fracture toughness of white cast irons through its ability to stop brittle cracks propagating from one carbide particle to another. Both abrasion resistance and fracture toughness were found to be greatest for austenitic matrices.

Cast steels made by electroslag refining have been shown to be capable of meeting the mechanical-property specification for forged gun tubes. Not only were the transverse ductility and impact toughness well within the specified minima, but also the degree of directionality between the transverse and longitudinal properties was much lower than for conventionally cast and forged gun barrels. An 81 mm dia. mortar made from an as-cast ESR ingot stood up well under the high impulsive loads imposed during firing tests with charges of up to 143% of normal.

An investigation into the effect of oxygen on the orthogonal cutting of copper, aluminium, and iron, using 40° rake-angled tools, has demonstrated that while reducing the partial pressure of oxygen increases the cutting forces with iron they are reduced significantly when cutting aluminium and copper. Experiments using internally marked copper–silver eutectic specinzens have shown that in the presence of oxygen there is a change in the geometry and type of secondary stress relief at the tool rake face from shear parallel to this face to intermittent twinning perpendicular to it throughout the thickness of the chip. That a similar process occurs in pure copper, only when cutting in oxygen, has been demonstrated by a comparison of the dislocation microstructure seen in three orthogonal sections of chips cut under differing environmental conditions.

The effect of iron on the activity of lead in the Cu–Fe–Pb solution containing 0·02–0·10 molfraction of iron and 0·02–0·06 molfraction of lead at 1473 and 1673 K was investigated. The activity of lead in the solution was measured by holding the Cu–Fe–Pb solution under investigation and a Cu–Pb standard solution with known thermodynamic properties, in a sealed crucible at a known argon pressure until equilibrium was established between the gas phases common to both solutions, and the solutions. In the equilibrium state the activity of lead in both solutions will be the same, according to Raoult's law. The effect of copper and iron vapours on the equilibrium was neglected because the rates of vaporization of these constituents in the experimental conditions are very low. The interaction parameter of iron and lead €Pb ^Fe in the Cu–Fe–Pb solution was determinedfrom the experimental results. It was found that over the concentration range examined, the Wagner linear equation describes accurately the experimental values of the activity coefficient of lead.

Strain-ratio values have been calculatedfor the following orientations of the tensilespecimen axis: 〈100〉, 〈10 10〉, 〈810〉, 〈510〉, 〈310〉, 〈210〉, 〈320〉, and 〈110〉. The results obtained made possible the determination of the distribution of the strain ratio in the {001} plane of an fcc metal. Theoretical results were verified in experiments on copper single crystals with the folio wing orientations: 〈100〉, 〈10 10〉, 〈510〉, 〈310〉, 〈210〉, 〈320〉, and 〈110〉. In each case, the single crystal was stepwise tensile tested and the strain ratio-true elongation (R–€) relationship was established, the extrapolation back to €=0 enabling the determination of the R ₀ coefficient. Good agreement between the calculated strain-ratio values and the experimental R ₀ coefficients was demonstrated. It was shown that an earlier prediction that the strain-ratio value for 〈110〉 orientation should be equal to zero was caused by the confusion of the strain-ratio value calculated at the limit of uniform elongation with its value obtained by extrapolation.

The historical development of the mechanically driven 4-hammer long forging machine is described. Comparisons are made of the operation and economic performance of the 4-hammer machine with conventional forging equipment, such as the hammer and hydraulic press. The 2-hammer long forging machine and its use to forge construction steels are outlined. Continuous forging machines, their use to forge continuously cast steel billet, and their combination with rolling stands for the production of bar steel are then described.

A CO₂ continuous wave laser producing up to 2 kW continuous power over an area down to 2·0 mm² was used as a surface heater. The surface hardening of certain steels is possible through a martensitic structure produced by rapid cooling from an elevated temperature, provided the laser power P, beam diameter D _b , and transverse speed V are kept within certain limits. Since the heat from a laser all passes into the surface and is not generated internally as with induction heating, thermal distortion is low and depth of hardening is controllable down to very thin layers. In the present paper the operating limits of the three principal process parameters when hardening En8 steel are described and it is shown that the depth of hardening is closely correlated with the parameter P/√D _b V. The resulting microstructures are analysed and the upper and lower speed limits for effective hardening discussed. Since the experiments were arranged statistically the reproducibility of hardness traces is also described.

The solidification process in some binary copper alloys was studied and special attention was devoted to the growth conditions of the primary phase (cells or dendrites). The actual growth temperature of crystals growing from the mould wall and inwards was measured as a function of constitution, composition, and rate of heat extraction and the morphology of the growing crystals was revealed by metallographic techniques on ‘frozen in’ samples. The formation and growth of ‘inner equiaxed’ crystals is closely coupled to the development of the columnar crystals and the relationship between the growth conditions of the different crystal types has been studied. In the same manner a number of commercial copper alloys were examined under varying casting conditions and the results referred to the findings from the binary alloys.

The correlation of non-homogeneity of texture in cold-rolled copper (99·9%purity) with that in recrystallized copper is described. A high degree of non-homogeneity of texture was obtained in the metal by a suitable rolling up to 90% deformation. The copper type of rolling texture appeared in the centre while the shear type was observed in the surface layer. By annealing between 700° and 900°C, a sharp cube texture and a small amount of texture similar to that observed in the deformed state were developed in the centre. The recrystallization type of texture appeared in the surface layer of the sample with an approximately 56° rotation about the common <111> directions of the suitable shear-texture components and was described as the {323}<101> orientation rotated by 7° about the transvere direction.

Specimens of 99·9 %purity copper were deformed initially by torsion up to €=200. Their mechanical properties were determined from tensile tests. The extreme values of mechanical properties were observed in a range of strain €=3−200. Changes in microstructure reflecting the mechanicalproperties of the specimens were investigated.