Architectured duplex stainless steels micro-composite: Elaboration and microstructure characterization

Architectured duplex stainless steels micro-composite: Elaboration and microstructure characterization

Materials and Design 145 (2018) 156–167 Contents lists available at ScienceDirect Materials and Design journal homepage:

6MB Sizes 0 Downloads 40 Views

Materials and Design 145 (2018) 156–167

Contents lists available at ScienceDirect

Materials and Design journal homepage:

Architectured duplex stainless steels micro-composite: Elaboration and microstructure characterization Hasan Naser a, Alexis Deschamps a,⁎, Marc Mantel a,b, Muriel Véron a a b

Univ. Grenoble Alpes, CNRS, Grenoble INP, SIMAP, 38000 Grenoble, France Ugitech Research Center, Avenue Paul Girod, 73403 Ugine Cedex, France




• For the first time a composite dual phase stainless steel has been produced by accumulative drawing and bundling. • The process enables to fabricate a dual phase microstructure while choosing the compositions of austenite and ferrite. • The process limitation occurs when inter-diffusion during annealing heat treatments provokes austenite reversion.

a r t i c l e

i n f o

Article history: Received 8 December 2017 Received in revised form 22 February 2018 Accepted 25 February 2018 Available online 26 February 2018 Keywords: Architectured duplex stainless steel Severe plastic deformation Micro-composites Accumulative drawing and bundling

a b s t r a c t In this work we propose a top-down strategy in which an austenitic stainless steel (type AISI 316L) and a ferritic stainless steel (type AISI 430LNb) are mechanically alloyed by Severe Plastic Deformation (SPD) to elaborate an architectured duplex stainless steel. This proposed strategy serves two main objectives: i) enhancing the properties by microstructure refinement down to sub-micron scale, and ii) elaborating a model material for understanding the behavior of Duplex Stainless Steel (DSS) obtained by the conventional metallurgical methods. The Accumulative Drawing and re-Bundling (ADB) technique has been successfully implemented for these specific materials, allowing to obtain multi-scale micro-composites of 316L/430LNb steels. The limits of this process in terms of microstructure refinement, have been identified as due to the complete regression of the micro-scale austenitic phase during annealing. © 2018 Elsevier Ltd. All rights reserved.

1. Introduction Combining two or more different materials in the view of obtaining a material with better properties than those of the individual components used separately is the basic concept of composite materials. In the late 1960s, a new class of metal-metal composite materials was developed, in which both matrix and the dispersed component can be codeformed plastically. Fe-Fe3C composites obtained by Embury and ⁎ Corresponding author. E-mail address: [email protected] (A. Deschamps). 0264-1275/© 2018 Elsevier Ltd. All rights reserved.

Fischer [1] showed a Hall-Petch type increase of strength with decreasing microstructure scale and their ultimate tensile strength reached 4.8 GPa. Although some authors do not classify Fe-Fe3C as a deformation processed metal-metal composite (DMMC) because of the limited plasticity of Fe3C, it has been considered as a precursor material for the development of DMMCs in various applications [2]. Depending on their manufacturing routes, these composites are obtained either by comelting two metals that are miscible as liquids but immiscible as solids followed by large section reduction using Severe Plastic Deformation (SPD) techniques to obtain filamentary structure called in-situ composites [2]; or by assembling mechanically the constitutive metallic

H. Naser et al. / Materials and Design 145 (2018) 156–167


Table 1 Nominal composition (in wt%) of the stainless steels used in the present work. Grade









430LNb 316L

0.014 0.019

18.05 16.73

0.2 11.10

0.039 2.034

0.382 0.795

0.016 0.029

0.490 0.028

0.398 0.505 Fig. 2. Configuration of the first composite.

components by SPD resulting in so-called co-deformed composites (or continuous composites) [3]. Among the latter category, there has been in particular a large interest for the accumulative roll bonding (ARB) process which has been applied to generate a large number of different composite systems, such as Al based [4–14], Mg-based [15,16], Cubased [17–22], Zr based [23], or Fe-based [24–29]. On the other hand, the interest in the accumulative drawing and bundling process (ADB) composites is to produce micro- or nano-structured wires. Cu-based composite wires have particularly been investigated in the aim to combine electrical conductivity and strength. The first deformation processed Cu-Nb in-situ composite obtained by [30] showed indeed an excellent combination of high strength-electrical conductivity. Following these results, several authors investigated Cu based composite with other metals such as Ag [30], Fe [31], Ta [32], V or Cr [33]. However, Cu-Nb is still considered as one of the most successful systems of mechanically alloyed composites using SPD techniques in terms of resulting properties and the comprehension of the deformation mechanisms. Nanocomposite wires of Cu-Nb obtained directly from the ADB process resulting in continuous fibers with controlled distribution of Cu and Nb, show interesting mechanical and electrical resistivity. Such nanocomposites have been investigated by [34–40]. Composites using other metals than Cu have been also explored e.g. for microelectronic applications such as Au-Ag [41] or more recently Al-Ca [42]. MetalMetal composites are also of interest for structural applications. For example, the systematic investigations realized on Al-Mg [43], Al-Sn [44] and Al-Ti [45] showed an ultimate tensile strength significantly higher than that of pure Al. Duplex Stainless Steels (DSS), with a microstructure consisting of approximately equal volume fractions of austenite and ferrite, have been recognized as interesting dual phase alloys for structural applications since the 1970s. DSS can be obtained by solidifying a specific chemical composition having a nickel content between 4 and 7%, and a chromium content of 18–25% [46]. They show good combinations of strength and ductility, such as a ultimate strength above 1GPa and a ductility of almost 25% for in-situ DSS composites designed recently

Fig. 3. SEM image in BSE mode of the microstructure of n1 composite in the as-drawn state (Ferrite: dark grey, Austenite: light grey).

[47,48], coupled with very interesting functional properties such as corrosion resistance and comparatively lower thermal conductivity. However, the mechanical properties of DSS are still the subject of many research investigations aiming at understanding their strength/ ductility compromise and at rationalizing microstructure/mechanical properties relationships. In [49] it has been proposed that the mechanical properties of DSS result from a complex interaction between ferrite and austenite leading to a mechanical behavior that cannot be predicted from the properties of the constituents alone, claiming that the strength

Fig. 1. a) Schematic illustration of the ADB process b) Evolution of the true deformation during drawing as a function of the step number where Si is the global input section and Sout is the global output section.


H. Naser et al. / Materials and Design 145 (2018) 156–167

Fig. 4. a) SEM images of the composites n1, n2, n3, n4 and n5 microstructures (dark grey = ferrite; light grey = austenite); b) n1 cells evolution inside composite n3, n4 and n5 after annealing at 850 °C for 1 h.

of DSS shows a significant deviation from the rule of mixture. The contribution of residual stresses and deformation incompatibility [50], caused mainly by microstructure morphology, were found to be unable to describe the mechanical behavior. Therefore, several authors used multi-scale modeling and in-situ experiments in order to analyze the load sharing and stress interaction between phases and consequently the micro stress-strain curves [51,52]. Nevertheless, this approach seems to be insufficient to explain the complex plastic interaction between ferrite and austenite [49].

Table 2 Dimension of each component inside composite n(i N 1).




Ferrite 430LNb Austenite 316L Ferrite 430LNb Austenite 316L

Ferrite 430LNb Austenite 316L


Ferrite 430LNb Austenite 316L

Issued from

Av. thickness (μm)

Volume fraction (%)

n1 n2 n1

79 114 104 28 34 39 37 11 – 117 8 10 132 10 5 – 117 2.5 3.6 37 2.2 1.5 – 11 152


n1 n2 n1 n2 n3 n1 n2 n4 n1 n2 n3 n1 n2 n4 n1 n2 n3 n5

Wire diameter Tube

Wire diameter Tube

Wire diameter Tube

Wire diameter Tube


The aim of the present study is to propose a different approach from that used until now to understand the behavior of DSS. This approach consists of designing an Architectured Duplex Stainless Steel (ADSS) starting from bulk austenite and ferrite stainless steels. Severe plastic deformation using ADB is well suited to this purpose since it offers freedom in terms of controlled parameters such as: chemical compositions of the constituent phases, phase distribution, scale, volume fraction etc. Moreover, this approach may not only allow understanding DSS behavior by creating model materials with controlled parameters but also developing new properties by microstructure refining. The purpose of the present study is to present processing conditions that allow for a successful processing of such ADSS by co-deformation, and to evaluate the limits of the process in terms of achievable microstructure scale.

2. Experimental procedure


2.1. Materials and processing


Two industrial grades provided by UGITECH, whose compositions are reported in Table 1, have been chosen as base materials: a ferritic stainless steel of type 430LNb and an austenitic stainless steel of type 316L stabilized by Nb and Ni, respectively [46]. Wires of diameter 1.55 mm and tubes with an external diameter of 6 mm and a thickness of 0.4 mm have been used in the current investigation. A single-pass Marshall drawing machine has been used to process the ADSS. It is composed of 4 parts; uncoiling, straighteners, die and its lubrication system, capstan. It allows a drawing velocity up to 3.3 m/s. The drawing velocity used in our experiments was between 0.5 and 1.5 m/s. A vacuum vertical furnace of capacity 850 mm ∗ 1500 mm was used for the annealing treatments. The maximum heating rate of this furnace is 70 °C/h. At the end of the heating cycle, a rapid cooling down to ambient temperature under argon atmosphere was operated once the vacuum was broken.





H. Naser et al. / Materials and Design 145 (2018) 156–167


Fig. 5. SEM micrographs showing the curly morphology of ferrite (dark grey) and austenite (light grey) in the as-drawn state of composite n5.

2.2. Characterization methods The composite wires were mainly observed in cross section. Metallographic preparation consisted in classical mechanical polishing with a 20 min Oxide Polishing Suspension (OPS) finish. Microstructural examination was first carried out using scanning electron microscopy using a LEO 440 SEM microscope operating at 20 kV. EBSD analysis has been

carried out with an automated pattern indexation tool (from TeX SEM laboratories) coupled to a FEG SEM Zeiss Ultra 55. EBSD data were post-processed using OIM software. The resolution of the investigated area depends on the step size. For large and local maps, a step size of 1 μm and 0.1 μm has been taken respectively. High-energy X-ray diffraction measurements were carried out in transmission mode at the ID15 beam line of the European Synchrotron

Fig. 6. EBSD maps showing the ferritic phase evolution: top, macroscopic maps for n3, n4 and n5 composites; bottom, local maps of n1 cells inside composite n3, n4 and n5: the austenitic phase is represented in dark. Yellow arrows in composite n3 and the blue circle in composite n4 show the presence of the small grains within the interface austenite/ferrite. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


H. Naser et al. / Materials and Design 145 (2018) 156–167

2.3. Composite fabrication 2.3.1. Methodology Architectured Duplex Stainless Steel (ADSS) wires are fabricated starting from individual wires of ferrite 430 LNb and austenite 316L, 1.55 mm in diameter, gathered in a containing tube of 430LNb or 316L (depending on the step number, see Section 3.2), 6 mm in external diameter and 0.4 mm in thickness. The fabrication involved the following successive steps as illustrated in Fig. 1: 0) Configuring the first composite starting from individual wires and tube. This composite constitutes the elementary pattern of the following composites structure. 1) Wire drawing: operation in which the (re-)bundled composite is reduced from φ = 6 mm to φ = 1.5 mm through 14 steps 2) Annealing: It consists on heat-treating the drawn composite in order to recover the material's ductility, which allows further drawing. 3) Re-bundling the annealed wire in a containing tube of ferrite or austenite.

The repetition of step 1, 2 and 3, namely accumulative drawing and bundling (ADB), produces filamentary micro-composites. We observed experimentally that a successful drawing in the above-mentioned conditions required the initial hardness of both materials to be below 220 HV. Trials with harder initial materials resulted in fracture during the drawing process. This value determined the temperature and duration of the annealing treatment for each step, which was limited by the annealing kinetics of austenite and fixed to 850 °C for 1 h, resulting in a hardness of 200 HV for austenite and 170 HV for ferrite, except for the first step where inter-diffusion was not an issue (in this case the annealing was realized at 1050 °C for 10 min). 2.3.2. Surface preparation Wires and tubes underwent a series of surface preparation steps at each generation of processing to manufacture the desired composites without any pollution:

Fig. 7. TEM-ACOM maps of 316L/430LNb interface in n4 composite before annealing within n1 cell: a) quality index map b) orientation map along drawing axis c) combined quality index and phase distribution map.

ESRF in Grenoble-France. The X-ray energy was 90 keV and the DebyeScherrer rings were acquired with a 2D camera. The data was radially averaged to obtain X-ray diffraction profiles and processed by Rietveld analysis using the Fulprof software. For fine-scale microstructure investigation, particularly the 316L/ 430LNb interface in high generation composites, JEOL 3010 and JEOL 2100F transmission electron microscopes (TEM) operating at 200 kV have been used. The thin foils were prepared using Focused Ion Beam (FIB). Grain orientation mapping in the TEM was performed using Automated Crystal Orientation Mapping (ACOM) using the ASTAR package [53]. Tensile tests on wires were carried out using specially designed grips to ensure that fracture did occur within the wire gauge length, at a strain rate of 2.10−4 s−1, with a wire length of 300 mm. The deformation was measured by an extensometer.

- Before bundling: the tube was coated by immersion in a hot bath of potassium sulphate (K2SO4) salt serving as a film support for lubrication whereas the wires were cleaned and degreased using an acetone bath to eliminate any surface contamination. - During drawing: two types of solid lubricant have been used to avoid a direct contact between the wire and the die; A calcium stearate with high filler amount (XCaHF) and a sodium stearate with low filler amount (XNaLF). - After steps of drawing: the wire surface was covered by a residual layer of coating and lubricant in the form of powder (soaps). Before subsequent heat treatment, it was necessary to remove this layer by pickling in a hot bath of phosphoric acid (H3PO4). - After annealing: despite the use of a vacuum furnace, some localized surface oxidation was observed. As mentioned above, before bundling, the wire was cleaned and degreased. In case of an oxidized wire, an operation of pickling with a hot bath (70 °C) of HCl was necessary. 3. Results 3.1. Composite fabrication The first composite consisted in 6 wires of ferrite 430LNb surrounding one wire of austenite 316L, these 7 wires being inserted inside an austenite tube as shown in Fig. 2. As this was the first step, the austenite and ferrite wires did not need a pickling operation. Only a degreasing operation was necessary.

H. Naser et al. / Materials and Design 145 (2018) 156–167


Fig. 8. TEM-EDS maps of 316L/430LNb interface in n4 composite before annealing within n1 cell; a) quality index map. Chemical concentration maps of b) Ni; c) Cr; d) Mo (dashed circle indicates precipitates).

Fig. 3 shows the resultant microstructure of the first composite (n1) in the as-drawn state i.e. after drawing without heat treatment. This composite was then heat treated at 1050 °C for 10 min in a vacuum vertical furnace. After annealing and surface preparation, the composite n1 was cut in seven wires of about 1 m length and inserted in either austenite or ferrite tube for the next generations of composite n(i N 1). After 92% of section reduction the dimension of wires and tube was divided by a factor 3.1 and the hardness increased significantly; from 200 HV to 450 HV for austenite and from 158 HV to 300 HV for ferrite. 3.2. Microstructural evolution Fig. 4 shows SEM images of the resultant microstructure of the composite n2, n3, n4 and n5. For generations n2 and n4, a ferrite tube was used as an outer jacket while an austenitic tube was used for generations n3 and n5. This allowed balancing the volume fraction to remain close to 50%γ-50%α during the iteration process. Table 2 summarizes the dimension of each component and the calculated volume fraction of both phases in each composite n (i N 1). In Fig. 4-a and -b, the presence of cracks can be noticed particularly in n(i-1) wires inside the composite n(i N 1) revealing the absence of bonding before and after annealing. However, these cracks were healed during the successive drawing steps. At a macroscopic level, the microstructure after drawing is homogeneous throughout the wire section, showing that the drawing strain was applied homogeneously. The microstructural evolution of the elementary patterns i.e. n1 cells inside composites n(i N 1) is shown in Fig. 4-b after annealing at 850 °C for 1 h. In composite n5, composed of 74 n1 cells, the dimension of the finest obtained phase is 1–3 μm. In this composite, n1 cells have no longer a cylindrical shape but start to get a wavy shape considered as the

signature of the curly structure (Fig. 5). A curly microstructure is a classical feature of BCC-FCC co-deformation [54]. This result is in agreement with the work of [55,56] on Cu-Fe co-deformation where it was shown that the curly structure begins when reaching a microstructure of about 3 μm. After a softening anneal of 850 °C for 1 h, the dual phase microstructure of the n5 composite and therefore its architecture is completely disrupted, as shown in Fig. 4-b. Indeed, one can observe an almost complete regression of the FCC phase (in dark) in particular the 2 μm austenite wire issued from n1 cells, thereby defining the limit of our refining process. The evolution of the phase distribution inside n1 cells of composites n3, n4 and n5 has been followed by EBSD. Fig. 6 presents EBSD crystallographic orientation maps where only the BCC phase is shown for clarity. After the annealing treatment the grain size of austenite (of the order of 3 μm) is significantly smaller than that of ferrite, which quickly becomes comparable to the size of the ferritic region in the composite. Aside the expected geometric distribution of phases resulting from the ADB process, one can observe the formation of small ferrite grains at the BCC/FCC interface, inside the formerly austenitic regions. These small grains are not visible in back-scattering SEM images where the chemical sensitive contrast assimilates them to the austenitic phase. Their presence increases significantly in the n4 composite where a complete regression of the FCC phase in favor of these small grains can be observed in some regions. In the n5 composite, EBSD map confirms the SEM observation showing an almost complete regression of the FCC phase and a disrupted architecture; the 2 μm austenite wires issued from n1 seem to be completely transformed into a BCC phase. For an in-depth understanding of this microstructural evolution, TEM observations have been performed on composite n4 within the vicinity of the interface α/γ of n1 cells. Using FIB preparation, thin foils


H. Naser et al. / Materials and Design 145 (2018) 156–167

found. From ACOM measurements, the thickness of this layer is found to be varying along the interface from 1 to 3 μm. Similarly to the asdrawn case, chemical analysis shows that this layer is richer in terms of Ni content than the ferritic phase. Moreover, similar precipitates as observed in the as-drawn state appeared in the as-annealed case (Fig. 10). 3.3. Texture The texture of each composite after annealing has been also followed using EBSD analysis (with a step size of 1 μm) on the cross section of the composite wire. Figs. 11 and 12 show the pole figures, where ND (drawing direction) is in the center, for the three components 〈100〉, 〈111〉 and 〈110〉 for both austenite and ferrite respectively. The austenite in the three composites develops a duplex fiber texture of 〈111〉 and 〈100〉. The distribution of these two components seems to be dependent on the composite. Indeed, as revealed by Fig. 11, composites n3 and n5 have an austenitic texture of 〈100〉 as major and 〈111〉 as minor component whereas the opposite trend is developed in composite n4. The ferritic phase develops a 〈110〉 fiber texture in the three composites. However, the intensity of this component seems to be also dependent on the composite; composite n5 shows a weaker intensity compared to composites n3 and n4. 4. Discussion 4.1. Microstructural evolution and mechanical properties

Fig. 9. TEM-ACOM maps of 316L/430LNb interface in n4 composite after annealing within n1 cell: a) quality index map b) orientation map along drawing axis c) combined quality index and phase distribution map.

were cut from n1 cells inside the n4 composite in both states, namely as-drawn and annealed, as illustrated in Figs. 7 and 9. The TEM samples were observed in the wire direction. Note that in Fig. 7 the map was realized on the sample edge so that the black area at the bottom of the image corresponds to the sample hole. In the as-drawn state, one can notice from Fig. 7a the presence of a thick layer within the vicinity of the large grains of ferrite, exhibiting an elongated lenticular structure. Thanks to automated crystal orientation mapping (ACOM), this lenticular structure is identified as mixed BCC-FCC phase. The thickness of this layer reaches almost 3 μm. Besides, TEM-EDS maps have been also acquired to analyze the distribution of chemical elements across 316L/ 430LNb interface and are shown in Fig. 8. The Ni content within this layer, and to a lesser degree the Mo content, is higher compared to that of the coarse grain ferritic phase. These maps reveal also the formation of some precipitates rich in Cr and Mo. These precipitates seem to have formed in the γ phase and in the vicinity of the formed layer. After annealing, the layer formed at the 316L/430LNb interface has been also identified and a contracted lenticular structure has been

The microstructure revealed by SEM images shows that for a given composite of range i (i = 1,2,3,4,5), the n(i-1) wires within this composite are not in complete adhesion even after annealing. This can be explained by the fact that the drawing was performed in 14 steps. Indeed, each pass has a total section reduction of at most 15% which is not sufficient to initiate a metallurgical bonding by plastic deformation only [57]. According to [57], if the threshold deformation is not reached during the first pass, it becomes too difficult to initiate bonding because the work hardening will increase further the subsequent threshold deformation. Hence, the metallurgical bonding between ferrite and austenite can only occur by diffusion during annealing if the two surfaces are in an intimate contact. It is for this reason that the n(i-1) wires inside a composite of range n(i) are not fully bonded even after annealing. However, this lack of bonding for the n(i-1) is healed by the subsequent drawing steps so that the n(i-2) wires are fully bonded. When the metallurgical bonding is achieved, all geometrical boundaries disappear and form a unique entity between ferrite/ferrite or ferrite/austenite components which means that the surface preparations and heat treatment in a vacuum environment were efficient so that neither inter-phase contamination nor oxidation were observed. EBSD and TEM observations showed that a specific microstructure is formed close to the interface α/γ on the formerly austenitic side, which means that part of austenite is transformed into BCC phase. This phase transformation can be directly attributed to the inter-diffusion during the successive annealing steps, in particular of Ni. During annealing the Ni diffusion from austenite (rich in Ni 11%) to ferrite (lean in Ni 0.2%) forms a depleted region near the α/γ interface to the detriment of the austenitic phase. This depleted region can produce either a new ferritic phase (BCC) having different chemical composition than the initial ferritic phase (430LNb) or an α′ martensitic phase upon cooling. The lenticular structure revealed by TEM observations before and after softening annealing confirms the formation of α′ to the detriment of the austenitic phase (316L). Indeed, the presence of α′ within the interface α/γ of n1 cells inside the composite n4 is inherited from the previous step (i.e. n3 annealed). Besides, for these annealing times and temperatures, the formation of other phases that might be found in stainless steels such as the σ-phase is not expected. It has been well established that for the stainless steels AISI grades used in the current work (316L and 430LNb), the precipitation kinetics of sigma phase is very slow

H. Naser et al. / Materials and Design 145 (2018) 156–167


Fig. 10. TEM-EDS maps of 316L/430LNb interface in n4 composite after annealing within n1 cell; a) quality index map. Chemical concentration maps of b) Ni; c) Cr; d) Mo (dashed circle indicates precipitates).

Fig. 11. Austenite texture inside composites n3, n4 and n5 after annelaing at 850 °C for 1 h.


H. Naser et al. / Materials and Design 145 (2018) 156–167

Fig. 12. Ferrite texture inside composites n3, n4 and n5 after annealing at 850 °C for 1 h.

and its formation requires thousands of hours [58–60]. The absence of a significant fraction of sigma phase in the fabricated microcomposites was confirmed by synchrotron X-ray diffraction measurements (see Fig. 13), and more locally by the ACOM-TEM maps. Therefore, the complete regression of the austenite of the n1 cells inside n5 composite generation is explained by the coupled effect of: i) the section reduction during drawing process (the austenite of n1 cells is reduced to 1–3 μm); ii) inter-diffusion during annealing, in particular Ni, between 316L and 430LNb and the martensitic transformation during cooling. Uniaxial tensile tests at ambient temperature were performed to explore the resultant mechanical behavior of the fabricated microscale composites. Fig. 14 shows the engineering stress-strain curves of composite n5 in as-drawn state (typical of all as-drawn composites) and those of composites n2, n3, n4 and n5 in the as-annealed state. As expected, the composites in the as-drawn state are strong but extremely brittle. Considering the effect of generation number on the mechanical properties, a systematic increase in yield strength has been found reaching 456 MPa in the as-annealed composite n5 with a uniform elongation of 19%. A detailed analysis of the mechanical behavior of ADSS is beyond the scope of this paper and will be the subject of forthcoming publication. In comparison to a classical DSS obtained by solidification such as UNS S32304 DSS, the yield strength of the multiscale microcomposite n5 is slightly increased as compared to 403 MPa for a coarse grain of UNS S32304 DSS in which 20 μm austenite phase is embedded in ferritic matrix [61]. However, its uniform elongation is decreased as compared to 26% [61].

4.2. Texture Ferrite developed the expected 〈110〉 fiber texture of the BCC metals recrystallized after an axisymmetric deformation. In the case of austenite, the texture evolution showed also the classical duplex fiber texture 〈100〉 and 〈111〉 of FCC metals. However, one can observe that the relative proportion 〈100〉/〈111〉 depends on the composite generation. Indeed, n5 and n3 showed the same fiber texture which is 〈100〉 as a major fiber, while n4 have 〈111〉 as a major component. This might be due to the thermo-mechanical history of the austenite. In fact, both composites n3 and n5 have an outer austenitic tube which constitutes almost 37% of the composite's microstructure. This tube underwent exactly the same process during the manufacturing of n3 and n5 composites i.e. drawing followed by annealing at 850 °C and 1 h as a holding time whereas in n4 composite, an outer ferritic tube was introduced and though the austenitic phase is only issued from n3 and n1 wires. 5. Conclusion The present study aimed at evaluating the potential of the Accumulative Drawing and Bundling (ADB) Process to elaborate a microscale composite from individual ferritic and austenitic stainless steels. The main results can be summarized as follows: - The ADB process has been successfully implemented for austenite/ ferrite stainless steel using intermediate annealing to restore the deformation capacity of the composite after each drawing step, and

H. Naser et al. / Materials and Design 145 (2018) 156–167


Fig. 13. X-ray diffractograms representing austenite (blue) and ferrite (red) peaks fitted with Lorentzian function for composites n2, n3, n4 and n5 after annealing at 850 °C for 1 h. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

micro-composite steels have been achieved. - The refining process is limited when the austenitic phase starts to regress significantly by inter-diffusion, resulting in a diminution of the fraction of austenite, and in a layer of martensite at the austenite/ferrite interface. - The as-drawn n5 composite is the finest microstructure obtained via this process in which micro-domains (1–3 μm) of austenite and ferrite are produced.

This ADB process applied to duplex stainless steels offers the potentiality to fabricate one-dimensional austenite/ferrite filamentary composites in which the composition of the two phases can be chosen, at least in the initial state, independently from the thermodynamic equilibrium conditions that prevail during the classical fabrication route of these materials.

Acknowledgements This work was performed within the framework of the Center of Excellence of Multifunctional Architectured Materials “CEMAM” n°AN-10LABX-44-01 funded by the “Investments for the Future Program”. F. Moser at Ugitech is gratefully acknowledged for help with the drawing process. N. Meyer is thanked for fruitful discussions. F. Robaut is thanked for help with EBSD observations. The technical staff of ID15 beamline at ESRF is also gratefully acknowledged for support during the experiments. Data availability The raw data required to reproduce these findings are available to download from the link 4w7w4hfdmk/draft?a=cd39a457-260a-40f7-9e29-aa562cb5a929. References Fig. 14. Tensile engineering stress vs. strain curves of the fabricated microcomposites.

[1] J.D. Embury, R.M. Fischer, The structure and properties of drawn pearlite, Acta Metall. 14 (1966) 147–159.


H. Naser et al. / Materials and Design 145 (2018) 156–167

[2] A.M. Russell, L.S. Chumbley, Y. Tian, Deformation processed metal–metal composites, Adv. Eng. Mater. 2 (2000) 11–22. [3] A. Bachmaier, R. Pippan, Generation of metallic nanocomposites by severe plastic deformation, Int. Mater. Rev. 58 (2013) 41–62, 1743280412Y.0000000003. [4] H. Pouraliakbar, A.H. Monazzah, R. Bagheri, S.M.S. Reihani, G. Khalaj, A. Nazari, M.R. Jandaghi, Toughness prediction in functionally graded Al6061/SiCp composites produced by roll-bonding, Ceram. Int. 40 (2014) 8809–8825, ceramint.2014.01.103. [5] A.H. Monazzah, R. Bagheri, S.M.S. Reihani, H. Pouraliakbar, Toughness enhancement in architecturally modified Al6061-5 vol.% SiCp laminated composites, Int. J. Damage Mech. 24 (2015) 245–262, [6] A. Anghelus, M.-N. Avettand-Fenoel, C. Cordier, R. Taillard, Microstructural evolution of aluminium/Al-Ni-Sm glass forming alloy laminates obtained by Controlled Accumulative Roll Bonding, J. Alloys Compd. 631 (2015) 209–218, 1016/j.jallcom.2014.12.231. [7] S. Khoramkhorshid, M. Alizadeh, A.H. Taghvaei, S. Scudino, Microstructure and mechanical properties of Al-based metal matrix composites reinforced with Al84Gd6Ni7Co3 glassy particles produced by accumulative roll bonding, Mater. Des. 90 (2016) 137–144, [8] F. Daneshvar, M. Reihanian, K. Gheisari, Al-based magnetic composites produced by accumulative roll bonding (ARB), Mater. Sci. Eng. B 206 (2016) 45–54, https://doi. org/10.1016/j.mseb.2016.01.003. [9] L. Ghalandari, M.M. Mahdavian, M. Reihanian, M. Mahmoudiniya, Production of Al/ Sn multilayer composite by accumulative roll bonding (ARB): a study of microstructure and mechanical properties, Mater. Sci. Eng.-Struct. Mater. Prop. Microstruct. Process. 661 (2016) 179–186, [10] M.M. Mahdavian, H. Khatami-Hamedani, H.R. Abedi, Macrostructure evolution and mechanical properties of accumulative roll bonded Al/Cu/Sn multilayer composite, J. Alloys Compd. 703 (2017) 605–613, [11] F. Kuemmel, H.W. Hoeppel, M. Goeken, Layer architecture and fatigue life of ultrafine-grained laminated metal composites consisting of different aluminum alloys, Mater. Sci. Eng.-Struct. Mater. Prop. Microstruct. Process. 702 (2017) 406–413, [12] R. Khatami, A. Fattah-alhosseini, Y. Mazaheri, M.K. Keshavarz, M. Haghshenas, Microstructural evolution and mechanical properties of ultrafine grained AA2024 processed by accumulative roll bonding, Int. J. Adv. Manuf. Technol. 93 (2017) 681–689, [13] P.D. Motevalli, B. Eghbali, Microstructure and mechanical properties of laminated Al-Cu-Mg composite fabricated by accumulative roll bonding, Bull. Mater. Sci. 40 (2017) 1481–1488, [14] A.H. Monazzah, H. Pouraliakbar, R. Bagheri, S.M.S. Reihani, Al-Mg-Si/SiC laminated composites: fabrication, architectural characteristics, toughness, damage tolerance, fracture mechanisms, Compos. Part B Eng. 125 (2017) 49–70, 1016/j.compositesb.2017.05.055. [15] G. Anne, M.R. Ramesh, H.S. Nayaka, S.B. Arya, S. Sahu, Microstructure evolution and mechanical and corrosion behavior of accumulative roll bonded Mg-2%Zn/Al-7075 multilayered composite, J. Mater. Eng. Perform. 26 (2017) 1726–1734, https://doi. org/10.1007/s11665-017-2576-z. [16] G. Anne, M.R. Ramesh, H.S. Nayaka, S.B. Arya, S. Sahu, Development and characteristics of accumulative roll bonded Mg-Zn/Ce/Al hybrid composite, J. Alloys Compd. 724 (2017) 146–154, [17] T. Nizolek, I.J. Beyerlein, N.A. Mara, J.T. Avallone, T.M. Pollock, Tensile behavior and flow stress anisotropy of accumulative roll bonded Cu-Nb nanolaminates, Appl. Phys. Lett. 108 (2016), 51903. [18] L.F. Zeng, R. Gao, Q.F. Fang, X.P. Wang, Z.M. Xie, S. Miao, T. Hao, T. Zhang, High strength and thermal stability of bulk Cu/Ta nanolamellar multilayers fabricated by cross accumulative roll bonding, Acta Mater. 110 (2016) 341–351, https://doi. org/10.1016/j.actamat.2016.03.034. [19] X. Ma, C. Huang, J. Moering, M. Ruppert, H.W. Hoeppel, M. Goeken, J. Narayan, Y. Zhu, Mechanical properties of copper/bronze laminates: role of interfaces, Acta Mater. 116 (2016) 43–52, [20] M. Hosseini, N. Pardis, H.D. Manesh, M. Abbasi, D.-I. Kim, Structural characteristics of Cu/Ti bimetal composite produced by accumulative roll-bonding (ARB), Mater. Des. 113 (2017) 128–136, [21] L.F. Zeng, R. Gao, Z.M. Xie, S. Miao, Q.F. Fang, X.P. Wang, T. Zhang, C.S. Liu, Development of interface-dominant bulk Cu/V nanolamellar composites by cross accumulative roll bonding, Sci. Rep. 7 (2017), 40742. [22] M. Alizadeh, M.K. Dashtestaninejad, Development of Cu-matrix, Al/Mn-reinforced, multilayered composites by accumulative roll bonding (ARB), J. Alloys Compd. 732 (2018) 674–682, [23] J.S. Carpenter, T. Nizolek, R.J. McCabe, M. Knezevic, S.J. Zheng, B.P. Eftink, J.E. Scott, S.C. Vogel, T.M. Pollock, N.A. Mara, I.J. Beyerlein, Bulk texture evolution of nanolamellar Zr-Nb composites processed via accumulative roll bonding, Acta Mater. 92 (2015) 97–108, [24] R. Jamaati, M.R. Toroghinejad, H. Edris, M.R. Salmani, Comparison of microparticles and nanoparticles effects on the microstructure and mechanical properties of steel-based composite and nanocomposite fabricated via accumulative roll bonding process, Mater. Des. 56 (2014) 359–367, 049. [25] R. Jamaati, M.R. Toroghinejad, S. Amirkhanlou, H. Edris, On the achievement of nanostructured interstitial free steel by four-layer accumulative roll bonding process at room temperature, Metall. Mater. Trans.-Phys. Metall. Mater. Sci. 46A (2015) 4013–4019, [26] K.S. Suresh, A.D. Rollett, S. Suwas, Evolution of texture and microstructure in deformed and annealed copper-iron multilayer, Metall. Mater. Trans.-Phys.




[30] [31]


[33] [34]

[35] [36]





[41] [42]




[46] [47]




[51] [52]

Metall. Mater. Sci. 47A (2016) 852–868, J.Q. Duan, M.Z. Quadir, W. Xu, C. Kong, M. Ferry, Texture balancing in a fcc/bcc multilayered composite produced by accumulative roll bonding, Acta Mater. 123 (2017) 11–23, H.R. Jafarian, M.F. Tarazkouhi, Significant enhancement of tensile properties through combination of severe plastic deformation and reverse transformation in an ultrafine/nano grain lath martensitic steel, Mater. Sci. Eng.-Struct. Mater. Prop. Microstruct. Process. 686 (2017) 113–120, 034. H.R. Jafarian, S.H.M. Anijdan, A.R. Eivani, N. Park, A comprehensive study of microstructure development and its corresponding tensile properties in nano/ultrafinegrained metastable austenitic steel during accumulative roll bonding (ARB), Mater. Sci. Eng.-Struct. Mater. Prop. Microstruct. Process. 703 (2017) 196–204, J. Bevk, J.P. Harbison, J.L. Bell, Anomalous increase in strength of insitu formed Cu-Nb multifilamentary composites, J. Appl. Phys. 49 (1978) 6031–6038. H. Brokmeier, Experimental textures of Al-Pb, Al-Cu and Fe-Cu metal-metal composites, Mater. Sci. Eng.-Struct. Mater. Prop. Microstruct. Process. 175 (1994) 131–139, W.A. Spitzig, A.R. Pelton, F.C. Laabs, Characterization of the strength and microstructure of heavily cold-worked Cu-Nb composites, Acta Metall. 35 (1987) 2427–2442, J.D. Verhoeven, F.A. Schmidt, E.D. Gibson, W.A. Spitzig, Copper-refractory metalalloys, J. Met. 38 (1986) 20–24, L. Thilly, F. Lecouturier, J. von Stebut, Size-induced enhanced mechanical properties of nanocomposite copper/niobium wires: nanoindentation study, Acta Mater. 50 (2002) 5049–5065, L. Thilly, V. Vidal, F. Lecouturier, Plasticity mechanisms in multi-scale copper-based nanocomposite wires, Mater. Sci. Forum 539–543 (2007) 814. V. Vidal, L. Thilly, F. Lecouturier, P.-O. Renault, Cu nanowhiskers embedded in Nb nanotubes inside a multiscale cu matrix: the way to reach extreme mechanical properties in high strength conductors, Scr. Mater. 57 (2007) 245–248, https:// J.B. Dubois, L. Thilly, F. Lecouturier, P. Olier, P.O. Renault, Cu/Nb nanocomposite wires processed by severe plastic deformation for applications in high pulsed magnets: effects of the multi-scale microstructure on the mechanical properties, IEEE Trans. Appl. Supercond. 22 (2012), 6900104. 2174574. L. Deng, X. Yang, K. Han, Y. Lu, M. Liang, Q. Liu, Microstructure and texture evolution of Cu-Nb composite wires, Mater. Charact. 81 (2013) 124–133, 1016/j.matchar.2013.04.013. T. Gu, O. Castelnau, S. Forest, E. Herve-Luanco, F. Lecouturier, H. Proudhon, L. Thilly, Multiscale modeling of the elastic behavior of architectured and nanostructured CuNb composite wires, Int. J. Solids Struct. 121 (2017) 148–162, 1016/j.ijsolstr.2017.05.022. T. Gu, J.-R. Medy, F. Volpi, O. Castelnau, S. Forest, E. Herve-Luanco, F. Lecouturier, H. Proudhon, P.-O. Renault, L. Thilly, Multiscale modeling of the anisotropic electrical conductivity of architectured and nanostructured Cu-Nb composite wires and experimental comparison, Acta Mater. 141 (2017) 131–141, j.actamat.2017.08.066. A. Russell, K. Xu, S. Chumbley, J. Parks, J. Harringa, A new method for strengthening gold, Gold Bull. 31 (1998) 88–92, L. Tian, A. Russell, T. Riedemann, S. Mueller, I. Anderson, A deformation-processed Al-matrix/Ca-nanofilamentary composite with low density, high strength, and high conductivity, Mater. Sci. Eng.-Struct. Mater. Prop. Microstruct. Process. 690 (2017) 348–354, K. Xu, A.M. Russell, L.S. Chumbley, F.C. Laabs, V.B. Gantovnik, Y. Tian, Characterization of strength and microstructure in deformation processed Al-Mg composites, J. Mater. Sci. 34 (1999) 5955–5959, K. Xu, A.M. Russell, Texture-strength relationships in a deformation processed Al-Sn metal-metal composite, Mater. Sci. Eng.-Struct. Mater. Prop. Microstruct. Process. 373 (2004) 99–106, A.M. Russell, T. Lund, L.S. Chumbley, F.A. Laabs, L.L. Keehner, J.L. Harringa, A highstrength, high-conductivity Al-Ti deformation processed metal metal matrix composite, Compos. Part - Appl. Sci. Manuf. 30 (1999) 239–247, 1016/S1359-835X(98)00163-8. K.H. Lo, C.H. Shek, J.K.L. Lai, Recent developments in stainless steels, Mater. Sci. Eng. R. Rep. 65 (2009) 39–104, C. Herrera, D. Ponge, D. Raabe, Design of a novel Mn-based 1 GPa duplex stainless TRIP steel with 60% ductility by a reduction of austenite stability, Acta Mater. 59 (2011) 4653–4664, H.-W. Yen, S.W. Ooi, M. Eizadjou, A. Breen, C.-Y. Huang, H.K.D.H. Bhadeshia, S.P. Ringer, Role of stress-assisted martensite in the design of strong ultrafine-grained duplex steels, Acta Mater. 82 (2015) 100–114, 2014.09.017. J.O. Nilsson, G. Chai, The physical metallurgy of duplex stainless steels, Int. Conf. Expo Duplex 2007 Grado Italy, Associazione di Italiana di Metallurgia 2007, pp. 73–82. I. Noyan, J. Cohen, X-Ray-diffraction study of changes in stress - strain distributions during the fatigue of a 2-phase alloy, Mater. Sci. Eng. 79 (1986) 149–155, https:// G. Chai, R. Lillbacka, Damage and crack initiation behavior of duplex stainless steel during cyclic loading, Key Eng. Mater. 324–325 (2006) 1117. N. Jia, R.L. Peng, Y.D. Wang, G.C. Chai, S. Johansson, G. Wang, P.K. Liaw, Interactions between the phase stress and the grain-orientation-dependent stress in duplex

H. Naser et al. / Materials and Design 145 (2018) 156–167

[53] [54]




stainless steel during deformation, Acta Mater. 54 (2006) 3907–3916, https://doi. org/10.1016/j.actamat.2006.04.019. E.F. Rauch, M. Veron, Automated crystal orientation and phase mapping in TEM, Mater. Charact. 98 (2014) 1–9, C.W. Sinclair, J.D. Embury, G.C. Weatherly, Basic aspects of the co-deformation of bcc/fcc materials, Mater. Sci. Eng.-Struct. Mater. Prop. Microstruct. Process. 272 (1999) 90–98, X. Quelennec, Nanostructuration d' un composite Cu-Fe par déformation intense: vers un mélange forcé à l'échelle atomiquePhD thesis University of Rouen, France, 2008. X. Quelennec, A. Menand, J.M. Le Breton, R. Pippan, X. Sauvage, Homogeneous Cu–Fe supersaturated solid solutions prepared by severe plastic deformation, Philos. Mag. 90 (2010) 1179–1195, L.R. Vaidyanath, M.G. Nicholas, D.R. Milner, Pressure welding by rolling, Br. Weld. J. 6 (1959) 13–28.


[58] V. Kuzucu, M. Aksoy, M.H. Korkut, M.M. Yildirim, The effect of niobium on the microstructure of ferritic stainless steel, Mater. Sci. Eng.-Struct. Mater. Prop. Microstruct. Process. 230 (1997) 75–80, 00010-5. [59] A.F. Padilha, D.M. Escriba, E. Materna-Morris, M. Rieth, M. Klimenkov, Precipitation in AISI 316L(N) during creep tests at 550 and 600 degrees C up to 10 years, J. Nucl. Mater. 362 (2007) 132–138, [60] H. Sahlaoui, H. Sidhom, Experimental investigation and analytical prediction of sigma-phase precipitation in AISI 316L austenitic stainless steel, Metall. Mater. Trans. - Phys. Metall. Mater. Sci. 44A (2013) 3077–3083, s11661-013-1647-5. [61] L. Chen, F.P. Yuan, P. Jiang, J.J. Xie, X.L. Wu, Simultaneous improvement of tensile strength and ductility in micro-duplex structure consisting of austenite and ferrite, Mater. Sci. Eng. A 618 (2014) 563–571,