As-quenched Structure Research Articles

Magnetic properties and structure of short-term annealed FeCuBPSi nanocrystalline alloys

Alloys with composition Fe80−xCu1B11P4Si4+x (x = 0 and 4), i. e. with Si 4 at % or 8 at % have been prepared by rapid quenching from the melt in air. For comparison the alloy with P 3 at % and Si 8 at % has been investigated as well. As-quenched structure of all ribbons was checked by X-ray diffraction. Subsequently the samples were submitted to thermal analysis, scanning and transmission electron microscopy and magnetic measurements. Hysteresis loops of strips after short-term annealing at 400, 420 and 500 °C for 5 min changed their shape – the saturation magnetic polarization increased. Its value changed from 1.67 to 1.74 T depending of the crystalline volume fraction for Fe80Cu1B11P4Si4 alloy. Addition of Si to the studied P 4 at % alloy leads to a decrease of coercivity after short-term annealing at the temperatures below the onset of the first transformation. Magnetic work decreased several times and its orientation acquired the distinct orientation along the ribbon axis.

The role of cementite on the hydrogen embrittlement mechanism in martensitic medium-carbon steels

Although cementite is the most common carbide found in steel components, its interaction with hydrogen (H) is still under discussion. Most studies investigate the effect of cementite by comparing the as-quenched vs. tempered structure of the same material. However, tempering reduces the dislocation density which plays an important role in several hydrogen embrittlement (HE) mechanisms. Therefore, this study investigates the HE behavior by in-situ hydrogen bending tests of two tempered medium-carbon martensitic steels with a similar dislocation density but a different cementite concentration. Moreover, a comparison between the as-quenched and the tempered samples is also made. This experimental strategy allows to generate important insights on the role of cementite on the susceptibility to HE while the influence of the dislocation density is limited. Thermal desorption spectroscopy and melt extraction results show that cementite acts as an effective H trap and reduces the H diffusion coefficient. When mechanically tested in a H rich environment, the presence of cementite increases the resistance to HE by hindering the H diffusion towards the most critical sites in the microstructure. However, when the samples are fully saturated with H prior to testing, cementite loses its capability to influence the fracture behavior.

Predicting the location of shear band initiation in a metallic glass

We report atomistic simulation results which indicate that the location of shear banding in a metallic glass (MG) can be ascertained with reasonably high accuracy solely from the undeformed static structure. Correlation is observed between the location of the initiation of shear bands in a simulated MG and the initial distribution of the density of fertile sites (DFS) for stress-driven shear transformations identified a priori based on a deep learning model devised in our recent work [Fan and Ma, Nat. Commun. 12, 1506 (2021)]. In addition, we demonstrated that one can judge whether a glass is brittle or ductile solely based upon its initial DFS distribution. These validate that shear bands in MG arise from non-linear instabilities, and that the as-quenched glass structure contains inhomogeneities that influence these instabilities. This work also reveals an important subtlety regarding the non-deterministic nature of athermal quasistatic shear simulations.

Nanocrystallization and Core-loss properties of Fe-rich FeSiBPNbCu nanocrystalline alloy

The present work investigates the structure, soft-magnetic and core-loss properties of 25 mm wide Fe83Si2B9P4Nb1Cu1 nanocrystalline ribbons. The melt-spun precursor ribbons of the thickness of 22 and 28 µm were prepared using commercial raw materials under ambient atmosphere. The XRD and DSC results show predominantly amorphous and hetero-amorphous structure for 22 and 28 µm ribbons in the as-quenched state. Further, the partial crystallization annealing leads to thickness dependent nanocrystallization process, wherein the thinner 22 µm ribbon shows sluggish primary(α-Fe(Si)) and delayed secondary(Fe3(B0.8P0.2)) crystallization process compared to 28 µm ribbons. The difference is explained through as-quenched precursor matrix structure and selective solute re-distribution of the intergranular region during nanocrystallization. Moreover, the higher crystallite size (D) and volume fraction(Vcr) of α-Fe(Si) nanocrystallites are observed for the 28 µm ribbon in the optimal annealing window. Under optimal annealing conditions, the 28 µm nanocrystalline ribbon (733 K) shows better AC soft-magnetic properties including B800 of 1.62 T, Hc of 15.7 A/m, Br/Bs ratio ≥ 0.8 compared to B800 of 1.59 T, Hc of 17 A/m and Br/Bs ratio ≥ 0.7 of 22 µm (743 K) ribbon. The 28 µm ribbon shows a low core loss (P) of 0.34 W/kg under 50 Hz, 1.5 T compared to P > 0.65 W/kg for 22 µm ribbons. The lower core-loss behaviour of thicker 28 µm in the sub-kHz frequency regime has been explained based on the nanocrystalline microstructure and loss coefficients (hysteresis and eddy current). The work compares the AC core-loss properties with reported Fe-rich nanocrystalline alloys and also discusses the scope of improving B800 beyond 1.65 T for the present alloys.

Reconstruction and Size Prediction of Prior Austenite Grain Boundary (PAGB) using Artificial Neural Networks

To automatically reconstruct the prior austenite grains from as-quenched martensitic structure, we applied a deep learning algorithm to recognize the prior austenite grains boundaries hidden in the martensitic matrix. The FC-DenseNet architecture based on FCN (fully convolutional networks) was used to train the martensite and ground truth label of the prior austenite grain boundaries. The original martensite structures and prior austenite grain boundaries were prepared using different chemical etching solutions. The initial PAGS detection rate was as low as 37.1%, which is not suitable for quantifying the basic properties of the microstructure such as grain size or grain boundary area. By changing the weight factor of the neural net loss function and increasing the size of the data set, the detection rate was improved up to 56.1%. However, even when the detection rate reached 50% or more, the quality of the reconstructed PAGS was not comparable to the analytically calculated results based on EBSD measurements and crystallographic orientation relationships. The prior austenite grain size data sets were obtained from martensite samples via the FCDenseNet method, and had a linear correlation with the mechanical properties measured in the same samples. In order to improve the accuracy of the detection rate using neural networks, it is necessary to increase the number of neural networks and data sets.

Crystallography and reorientation mechanism upon deformation in the martensite of an α-α’ Ti-6Al-4V dual-phase microstructure exhibiting high work-hardening rate

The present study provides a fundamental understanding of the crystallography and the microscale behavior of the V-enriched and Al-depleted α’ martensite taking place during the microstructure formation and the deformation of the dual phase α + α’ microstructure produced in Ti-6Al-4V. This particular microstructure exhibits much larger work-hardening capabilities than the conventional wrought product. The as-quenched structure of the martensite is analyzed using the Phenomenological Theory of the Martensite Crystallography (PTMC) coupled with EBSD analyses. This approach sheds new light on the microstructure configuration obtained during the dual-phase treatment and its fine-scale mechanical behavior. The martensite preferentially forms into parallel groups of 3 self-accommodating variants separated by a misorientation of 63.26∘/[10¯553¯]α’. TEM analyses additionally show that the variants of a same group are separated by a hitherto unobserved {134¯1}α’ type twin plane. Postmortem analyses after tensile testing demonstrate that this twinning plane is mobile under deformation. This allows the martensitic microstructure to exhibit the remarkable property to reorient under uniaxial tension. This unique property is shown to be intimately related to the mobility of the {134¯1}α’ twinning plane thereby evidencing for the first time that twin boundary motion is not uniquely associated to the orthorhombic α’’ martensite but can also occur in hexagonal α’ martensite. Quantification of the Interaction Energy (IE) appears relevant to rationalize and predict the reorientation of the martensite. The critical influence of the parent β grains texture on the reorientation is evidenced, while the impact of this deformation mechanism on the ductility of the martensite is debated.

Hierarchical Cu precipitation in lamellated steel after multistage heat treatment

The hierarchical distribution of Cu-rich precipitates (CRPs) and related partitioning and segregation behaviours of solute atoms were investigated in a 1.54 Cu-3.51 Ni (wt.%) low-carbon high-strength low-alloy (HSLA) steel after multistage heat treatment by using the combination of electron backscatter diffraction (EBSD), transmission electron microscopy (TEM) and atom probe tomography (APT). Intercritical tempering at 725 °C of as-quenched lathlike martensitic structure leads to the coprecipitation of CRPs at the periphery of a carbide precipitate which is possibly in its paraequilibrium state due to distinct solute segregation at the interface. The alloyed carbide and CRPs provide constituent elements for each other and make the coprecipitation thermodynamically favourable. Meanwhile, austenite reversion occurs to form fresh secondary martensite (FSM) zone where is rich in Cu and pertinent Ni and Mn atoms, which gives rise to a different distributional morphology of CRPs with large size and high density. In addition, conventional tempering at 500 °C leads to the formation of nanoscale Cu-rich clusters in α-Fe matrix. As a consequence, three populations of CRPs are hierarchically formed around carbide precipitate, at FSM zone and in α-Fe matrix. The formation of different precipitated features can be turned by controlling diffusion pathways of related solute atoms and further to tailor mechanical properties via proper multistage heat treatments.

The effect of carbon content on the c/a ratio of as-quenched martensite in Fe-C alloys

The as-quenched martensite crystal structure is widely accepted as body-centered tetragonal. The classical Honda & Nishiyama model, c/a = 1 + 0.045wt%C, was obtained based only on the experimental results for steels containing more than 0.6wt% of carbon. This model was used to predict that steels containing less than 0.6wt% of carbon would follow the same relation. However, it was reported by Sherby that the steel with less than 0.6wt% of carbon is body-centered cubic, and c/a ratio equals to 1. This debate is caused by highly overlapped martensite (002) and (200) peaks when the carbon content of steel is less than 0.6wt%. In this paper, the martensite lattice parameters of the as-quenched steels with different carbon contents were measured using high-resolution X-ray diffractometer. Rietveld refinement was used to deconvolute overlapped martensite peaks. For the steel with less than 0.6wt% of carbon, the structure is proved to be body-centered tetragonal. It was found that the relationship between c/a ratio and carbon content follows as c/a = 1 + 0.031wt%C.

Tempering of Lath Martensite in Low and Medium Carbon Steels: Assessment and Challenges

This paper is a directed review of the microstructures and the mechanisms, by which they are formed by tempering of lath martensite in low and medium carbon steels. As-quenched martensitic structure is first reviewed, then low-temperature tempering (LTT) that produces ultrahigh strength and good toughness, and high-temperature-tempering (HTT), which can produce excellent combinations of strength, ductility, toughness, and good resistance to hydrogen embrittlement in sour gas and oil environments, are covered. It is well known that hardness decreases continuously with increasing tempering temperature, but the superimposed multiple mechanisms of microstructural changes by which the softening and its retardation occur during various stages of tempering have received little attention on the integrated scales of microstructure, substructure, and nanostructure. Thus, earlier descriptions of the reasons for softening need deeper examination. In particular, the reasons for the changes in the very fine crystal size and high dislocation density of as-quenched lath martensite produced by high temperature tempering, sometimes producing a low dislocation density, un-recrystallized, fine lath-morphology ferritic grain size, sometimes producing an equiaxed, recrystallized ferritic grain size, are not clear. Deformation behavior and mechanical properties as related to the residual components of martensitic microstructure that survive tempering are discussed. Questions remain and the use of newer techniques, such as Electron Back Scatter Diffraction and atom probe tomography, in addition to light microscopy, transmission electron microscopy, Mossbauer effect spectroscopy, and Mechanical Spectroscopy used in characterizing quench and tempered structures are noted.

High entropy effect on structure and properties of (Fe,Co,Ni,Cr)-B amorphous alloys

We examined high entropy (HE) effect on the formation, structure, thermal stability, mechanical properties and corrosion behavior for (Fe,Co,Ni,Cr)100-xBx (x = 18–26) amorphous alloys in comparison with Fe100-xBx amorphous alloys. The as-quenched structure consists of an amorphous single phase for the 18–24%B alloys, and amorphous + tetragonal Cr5B3 phases for the 26%B alloy. The broad diffraction peak position due to amorphous phase shifts to a higher wave vector side with increasing B content for the HE alloys, while that of the Fe-B alloys changes to the lower wave vector side, indicating the distinct difference in disordered atomic structure between the HE alloys and the binary alloys. All the as-spun HEA ribbons exhibit good bending ductility in spite of the coexistence with their crystalline phases, being different from the brittle nature for the 26%B binary alloy. The crystallization temperature (Tx) and Vickers hardness (Hv) increase with increasing B content, show maximum values of 803 K and 1130, respectively, at 22%B and then decrease to 793 K and 1070, respectively at 26%B, being different from the nearly constant Tx and almost linear increase in Hv with increasing B content for the Fe100-xBx alloys. The corrosion resistance in NaCl, H2SO4 and HCl solutions evaluated by potential and anode current density is much better for the HE alloys as compared with the Fe-B alloys. Thus, the Tx, Hv, bending ductility and corrosion resistance for the HE amorphous alloys containing more than 20 at%B are much superior to those for the Fe-B amorphous alloys. The fully crystallized structure consists of fcc + FeB + Fe2B + Cr5B3 phase for the HE 22–26 at%B alloys and bcc-Fe + Fe2B for the Fe-B alloys. The significant improvement of the fundamental properties caused by the HE component modification is presumably due to the formation of Cr-B pair with higher B-rich components. The improved properties in conjunction with the change in the disordered structure for the HE amorphous alloys provide an opportunity of developing a new functional material by alloy design to HE type compositions.

The role of amorphous precursor in phase selection hierarchy in marginal metallic glasses

The solid state amorphous structure and devitrification products of Al90Tb10 alloys produced by Cu block melt-spinning and magnetron sputtering were investigated by a combined study of differential scanning calorimetry, in-situ X-ray diffraction and conventional transmission electron microscopy. The as-prepared specimens were found to be fully amorphous according to the electron and X-ray diffraction. The thermal analysis and in-situ XRD analyses of the amorphous samples having the same chemistry but different processing history showed clear differences in the devitrification pathways in terms of nanocrystal population, size and the phase selection. The variations in the crystallization path were explained on the basis of distinct topological and chemical medium range ordering observed in as-quenched and as-sputtered amorphous structures.

Microstructural Features Affecting Tempering Behavior of 16Cr-5Ni Supermartensitic Steel

In industrial production processes, the respect of hardness and UTS maximum values of 16Cr5Ni steel is of utmost importance and a careful control of chemical composition and thermo-mechanical treatments is a common practice. Nevertheless, some scatter of properties is often observed with consequent rejection of final components. To better understand the role played by different factors, two heats of 16Cr-5Ni supermartensitic stainless steels with very close chemical compositions but different thermal behavior during tempering have been studied by means of TEM observations, X-ray diffraction measurements, dilatometry, and thermo-mechanical simulations. It has been found that Ms–Mf temperature range can extend below the room temperature and the relative amount of retained austenite in as-quenched conditions plays a significant role in determining the thermal behavior. When present, the γ-phase increases the amount of reversed austenite formed during tempering and accelerates the process kinetics of martensite recovery. Moreover, increasing amounts of retained austenite after quenching lower the critical temperature for austenite destabilization and influence the optimum temperature–time combination to be adopted for controlling final mechanical properties. In the studied cases, the very close chemical composition of the heats was not a sufficient condition to guarantee the same as-quenched structure in terms of retained austenite amount. This was proven to be related to solute segregation effects during solidification of original heats.

Influence of Ti addition on the microstructure and mechanical properties of a 5% Cr–Mo–V steel

A systematic study to determine the effect of Ti addition on the microstructure and mechanical properties of a 5% Cr–Mo–V tool steel has been made in terms of optical and electron microscopy, as well as X-ray diffraction, thermal analysis, and small angle neutron scattering. With the characterization of the as-quenched structure, the microstructural variation as a function of aging time was investigated. The Ti addition effectively refines the prior austenite grain structure as expected. More importantly, it can change the precipitation behavior of secondary carbides formed upon aging. As this change is attributed to the resisted Cr diffusion into cementite, the resultant precipitation of M23C6 carbides was restricted. Hence this leads to a significant improvement of the hardness to toughness balance, which is coupled with grain refinement. This variation in precipitation behavior was considered to have a thermodynamic aspect, especially the driving force for nucleation of equilibrium carbides involved in the present alloy system.

Modifier cation (Ba, Ca, La, Y) field strength effects on aluminum and boron coordination in aluminoborosilicate glasses: the roles of fictive temperature and boron content

The field strength of modifier cations has long been known to have important effects on oxide glass properties, but effects on network structure can be complex. For two series of barium, calcium, lanthanum and yttrium aluminoborosilicates with two different B/Si ratios, we report systematic variations in boron and aluminum coordination determined by NMR, and glass transition and heat capacities from differential scanning calorimetry. Data on glasses with different fictive temperatures allow B and Al speciation to be compared on an isothermal basis, rather than as conventionally done for as-quenched structures. Temperature and compositional effects can thus be isolated. These data and comparison to previous studies on glasses with lower B/Si ratios clearly show that higher modifier cation field strength increases the fraction of five- and six-coordinated Al in all compositions. In contrast, the previously documented trend towards more three-coordinated boron (and hence more non-bridging oxygens, NBO) in low B/Si glasses with higher field strength cations reverses in high B/Si and in high NBO compositions. Al and B coordination numbers both decrease with higher fictive temperature in the glasses studied here, suggesting a simple mechanism of coupled structural change.

Designing Cu-Zr Glass Using Multiobjective Genetic Algorithm and Evolutionary Neural Network Metamodels–Based Classical Molecular Dynamics Simulation

Shear deformation analysis and diffusion behavior of Copper-Zirconium Bulk Metallic Glasses (BMG) is studied for different combinations of processing parameters. Melt holding temperature, melt holding duration, cooling rate of melt and composition of BMG are varied to obtain BMGs of different structures. The as-quenched structures are characterized using Radial Distribution Function (RDF) and Self-Diffusion constant values (of each atom type). The objective of this study is to design a Cu-Zr BMG which can absorb maximum energy and still deform as little as possible while maximizing its diffusivity during shear deformation of the structure. For this, metamodels were constructed by feeding the Molecular Dynamics (MD) results to an Evolutionary Neural Network (EvoNN) so as to generate the desired objective functions which are then optimized through multiobjective genetic algorithm. This led to identification of some hitherto unknown structures, characterized by atomic coordinates, which have good resistance to shear deformation and at the same time possess high diffusivity.

Combination of TEM and 3D atom probe study of microstructure and alloy carbide in a Nb-V micro-alloyed steel tempered at 650 °C

3D atom probe (3DAP) combined with TEM is applied to characterize the nanostructure and composition of coarsening carbides that precipitate in a Nb-V micro-alloyed steel tempered at 650 °C for 4 h. The results indicate that high temperature tempering of the as-quenched lath-like martensitic structures leads to the coarsening of alloyed carbide in the recovered ferritic matrix. TEM characterization and EDS analysis show that the alloy carbides are typically spherical or ellipsoidal and are enriched with V, Nb and Mo. Crystallographic lattice overlap and distortion create a distinct optical Moire pattern at the carbide/matrix boundary, as observed by HRTEM. 3DAP gives the compositional distribution at atomic scale in the coarsening carbide and the surrounding matrix. Non-carbide-forming elements such as Si and Al are depleted in the carbide and are enriched at the carbide/matrix heterophase interface. Carbide-forming elements such as Mn, V, Mo and Nb distribute heterogeneously in the whole carbide, leading to the formation of a Mn-, V- and Mo-enriched core and Mo-enriched external shell.

New Fe-B-P-Cu Nanocrystalline Soft Magnetic Alloys with High <i>J</i><sub>s</sub> Combined with Low Coercivity <i>H</i><sub>c</sub>

Fe-Si-B Amorphous Alloys with Less than 80 at% Fe Are now in Practical Use due to their Excellent Magnetic Softness (Low Coercivity Hc) Combined with Rather High Saturation Magnetic Polarization (Js) which Basically Owing to the Lack of Intrinsic Magnetic Anisotropy and the High Fe Content, Respectively. In Order to Obtain High Js, High Fe Content Is Required. However, Alloys with High Fe Content Exceeding the Limit Usually Have the as-Quenched Structure Consisting of Coarse α-Fe Grains in the Amorphous Matrix, which Results in Inferior Magnetic Softness. We Have Developed a New Fe85.2B10P4Cu0.8 Nanocrystalline Soft Magnetic Alloy Ribbon (with 5 mm in Width and about 20 µm in Thickness) Made from Industrial Raw Materials in Air Atmosphere. The as-Quenched Structure of Fe85.2B10P4Cu0.8 Alloy Has Heterogeneous Amorphous Structure (a Large Amount of Extremely Small α-Fe Clusters in Addition to Amorphous Phase). Homogeneous Nanocrystalline Structure Composed of α-Fe Grains with a Size ~19 nm Was Realized by Crystallizing the Hetero-Amorphous Alloy. The Nanocrystalline Alloy Exhibit High Js ~ 1.83 T (Comparable to the Commercial Fe-3.5 Mass% Si Steel) and Extremely Low Hc ~ 6.0 A/m. Additionally the Alloy Has a Large Economical and Industrial Advantage of Lower Material Cost and Good Reproductivity, which Has a High Potential for the Power Applications.

Effect of Austenitizing Heat Treatment on the Microstructure and Hardness of Martensitic Stainless Steel AISI 420

The effect of austenitizing on the microstructure and hardness of two martensitic stainless steels was examined with the aim of supplying heat-treatment guidelines to the user that will ensure a martensitic structure with minimal retained austenite, evenly dispersed carbides and a hardness of between 610 and 740 HV (Vickers hardness) after quenching and tempering. The steels examined during the course of this examination conform in composition to medium-carbon AISI 420 martensitic stainless steel, except for the addition of 0.13% vanadium and 0.62% molybdenum to one of the alloys. Steel samples were austenitized at temperatures between 1000 and 1200 °C, followed by oil quenching. The as-quenched microstructures were found to range from almost fully martensitic structures to martensite with up to 35% retained austenite after quenching, with varying amounts of carbides. Optical and scanning electron microscopy was used to characterize the microstructures, and X-ray diffraction was employed to identify the carbide present in the as-quenched structures and to quantify the retained austenite contents. Hardness tests were performed to determine the effect of heat treatment on mechanical properties. As-quenched hardness values ranged from 700 to 270 HV, depending on the amount of retained austenite. Thermodynamic predictions (using the CALPHAD™ model) were employed to explain these microstructures based on the solubility of the carbide particles at various austenitizing temperatures.

Influence of Structure Evolution on Magnetic Properties of Fe–Ni–Nb–B System

The effect of combined presence of iron and nickel has been studied in rapidly quenched amorphous (Fe-Ni) 81Nb7B12 system with the ratio of Fe/Ni=2/1 and 1/2 in as-quenched state and after annealing. Field dependencies of magnetostriction as well as the values of saturation magnetostriction were correlated with the evolution of nanocrystalline structure in amorphous matrix in the temperature range from ~ 700 to 800 K and after complete crystallization above 900 K. Intervals of stability and transformation regions were determined from temperature dependencies of electrical resistivity. The structure after annealing at selected temperatures was identified by X-ray diffraction (XRD), transmission electron microscopy (TEM), and electron diffraction. MO?ssbauer spectroscopy was used only as a complementary method to demonstrate the behaviour and development of the paramagnetic (FeNi) 23B6 phase from the original as-quenched structure. MO?ssbauer and magnetostriction measurements were performed at room temperature. The observed field dependencies of magnetostriction are a combination of magnetostrictions of ferromagnetic and paramagnetic phases formed during the transformation process, namely, the nanocrystalline cubic Fe-Ni phases and face-centered cubic (fcc)-type structure (FeNi)23B6, as identified from the structure analyses. The transition from purely ferromagnetic to partially paramagnetic state is well observed in the evolution of MO?ssbauer spectra evolution. Different ratios of both ferromagnetic components Fe and Ni lead to a change of the structure of the nanocrystalline phases and thus also to a change in the magnetic behavior of the system.

Precipitate assemblies formed on dislocation loops in aluminium-silver-copper alloys

The precipitation microstructure of the γ′ (AlAg2) intermetallic phase has been examined in aluminium-silver-copper alloys. The microstructure developed in an Al-0.90at.%Ag-90at.Cu alloy was significantly different from that reported for binary Al-Ag alloys. The orientation relationship between the matrix and precipitate was unchanged; however, the γ′ phase formed assemblies with a two-dimensional, open arrangement of precipitates. Each such assembly contained two variants of the γ′ phase alternately arranged to form a faceted elliptical unit. The θ′ (Al2Cu) phase formed on these assemblies after further ageing. Each assembly was formed via repeated precipitation of the γ′ phase on dissociated segments of a single dislocation loop. This faceted elliptical assembly has not been previously reported for the γ′ precipitate. The difference between the precipitation behaviour of the γ′ phase in Al-Ag and Al-Ag-Cu alloys was attributed to copper modifying the as-quenched defect structure of the matrix. The formation of faceted elliptical γ′ phase assemblies clarifies earlier observations on the precipitate number density and mechanical properties of aluminium-silver-copper alloys.